Network Economics – for Lo205 Judith Molka-Danielsen j.molka-danielsen@himolde.no http://home.himolde.no/~molka Reference: “Information Rules”, Carl Shapiro and Hal R. Varian, Havard Business Sch. Press, chapter 07, 1999. March 01, 2002

Overview Describe the New Information Economy Discuss the factors in the Economics of Networks Define the terms: network effects and network externaltities Discuss the concept of Lock-in

The old industrial economy Dominated by Oligopolies a few large firms market share rose and fell slowly lifetime employment Industries that exhibited this: automobile steel aluminum petrolium chemical markets economies of scale

The new information economy Dominated by temporary monopolies dominate player based on the days leading technology short time on top 7 years average at one job Industries: hardware and software economies of networks

Types of Networks Physical networks High Technology (real) telephone, electrical, railroad, airlines High Technology (real) computer, fax machine, compatible modems, email, ATMs, Internet High Technology (virtual) Macintosh users, CD readers-writers users, Nintendo 64 users, minidisc users

Important Concepts It is better to be connected to a bigger network than to a smaller one. Network effects Positive Network Externalities Companies compete by expanding their networks, they increase the value by interconnecting with other networks.

Positive Feedback Positive feedback - the strong get stronger and the weak get weaker. Speaking into a microphone, when the amplified signals go back into the system, you have repeated amplification. It gets loud. Negative feedback - when the strong get weaker and the weak get stronger. Go to a middle point. Industrial oligopolies exhibit. In the industrial markets, attempts for greater market share would trigger responses by others towards efficiency, also too large firms were too complex to manage. The norm were oligopolies.

Positive feedback (cont.) Postitive feedback can contribute to growth, but it is not the same thing. Ie. Internet. Virtuous cycle - success feeds itself. Vicious cycle - product seen as failing, the perception feeds the cycle, death spiral, the weak get weaker. Ie. Apple Macintosh. Tippy market - two players compete, one wins. (Video recorders - VHS v. Beta) Winner take all.

Positive Feedback - example Nintendo - enters home video games in 1985 Other dominant player at the time (Atari, Texas Instruments) Christmas 1986, Nintendo Entertainment System (NES) was hot. More game developers begin to write games for them. Game developers pay royalties to Nintendo and agree to not share the game with other systems for 2 years after release.

Sources of Positive Feedback (V&S) Supply side economies of scale Declining average cost Marginal cost less than average cost Example: information goods Demand side economies of scale Network effects In general: fax, email, Web In particular: Sony v. Beta, Wintel v. Apple

Chicken & Eggs (V&S) Fax and fax machines VCRs and tapes Internet browsers and Java

S-shaped curve Adoption of new technologies follow an S-shaped curve in three phases: flat during launch, steep rise during takeoff, positive feedback at work, leveling off, saturation is reached. Seen in the adoption of: fax machine, CD, color TV, video games, e-mail, and the Internet.

Economies of Scale Economies of scale - is a source of positive feedback. Larger firms tend to have lower unit costs of productions. This is supply-side economies of scale. In manufacturing the benefits of scale are usually exhausted before total market dominance. Demand side economies of scale - Microsoft Windows 95 in 1998, widely used, de facto industry standard. Some technologies go head-to-head for years before a winner emerges. Telephone netwk also showed this.

Network externalities Sponsor - of a net builds and hopes to profit from the net. Apple controls interfaces, clone license terms, architecture improvements, for the Mac. And influence supply of complementary products. (Partners) Recall externalities are positive and negative (pollution) Positive network externalities lead to positive feedback Metcalfe’s law: The value of the network goes up as the square of the number of users. Total value of the network to all users = n x (n-1) = n2 -n where n is the number of users. If value=$1 for a single user, and n=10, then network size 10 has 102-10=90 = $90.

Collective switching costs Switching costs - come from complementary assets, hardware with software, IT systems with training. If I learn to program in Access db language other Access software and the language becomes more valuable to those who are already users. But it is hard to get people to switch, collective switching costs. It is hard to coordinate users to switch all at the same time, and they will not do so by themselves. Therefore, the existing providers have the advantage. 1870 QWERTY keyboard wins over the 1932 Dvorak layout with home row AOEUIDHTNS that was proven to be faster. We would collectively be better to switch but the individual switching costs are too high. (Also called technological Lock-in.)

Summary on Lock-In and Switching Costs (V&S) Network effects lead to substantial collective switching costs Collective lock-in is even worse than individual lock-in Due to coordination costs Example: QWERTY

Don’t Get Carried Away (V&S) Network externalities don’t always apply ISPs (but watch out for QoS) PC production (easy to produce)

Industry and positive feedback Not every market tips (Ford had a best-selling-car, but who cares, you do not buy a car because others have it. But you may avoid a car because no one has it.) There are no demand side economies of scale within the IBM compatible PC market because standards allow interoperability. Another example, Internet service providers. (America Online, CompuServe, Delphi, had separate discussion groups before Internet. Now access can be from anyone, AOL, IBM, ATT, etc.) ISPs could differentiate access through QoS offers, video conferencing, etc. Large ISPs have advantage because it is easier to control QoS on a single network.

Likelihood of a market tipping to one technology *sum of economies of scale - ie. There are supply side eos for PC market where Compaq, Dell, HP and IBM have 24%. **standards reduce variety, but global HDTV standards are not needed.

Performance v. compatibility How do you start the virtuous cycle? Philips and Sony introduced the CD in the 1980s. Philips offers digital compact cassette (DCC) in 1992 where DCC players can play regular cassettes, backward compatible. But it didn’t catch on. Consumer inertia approaches: the evolution strategy of compatibility, and the revolution strategy of performance. Revolution - best performance to start over with a new product, but high switching costs. Evolution - give up some performance to allow compatibility (lower switching costs).

Igniting Positive Feedback (V&S) (First fundamental tradeoff) Evolution Give up some performance to ensure compatibility, thus easing consumer adoption Revolution Wipe the slate clean and come up with the best product possible

Evolution (V&S) Offer a migration path Examples Microsoft Intel Borland v Lotus Build new network by links to old one Problems: technical and legal

Technical Obstacles to Evolution (V&S) Use Creative design Think in terms of system Converters and bridge technologies One-way compatibility

Legal Obstacles (V&S) Need IP licensing Example: Sony and Philips CDs

Openness v. Control Is the second fundamental tradeoff. Open - make interfaces and specifications available to others. Consumers can fear lock-in, open raises chances of success, attracks allies. Control - keep your system proprietary. (This is valuable if your system takes off.) Installed base is more valuable if rivals cannot offer products to lock-in customers. Control interconnection. Existing market position, technical capabilities and control of intellectual property rights are 3 dimensions of critical market strength. Intel controls MMX multimedia specs for its Pentium chips, but promote an open interface spec for graphic controllers, the accelerated graphics port (AGP) to increase demand for their microprocessors.

Openness v. Control (V&S) Your reward (total value to you) = Total value added to industry x your share (in the market). Total Value added to Industry (VI) Depends on product and Size of network VI= the value the technology adds to the existing product value + how widely the technology is adopted (network size). Your share (YS) of industry value Depends on how open YS= your market share + your profits + your royalties + how the technology effects your sale of other products (add to sales or take away).

Openness (V&S) Full openness Alliance Anybody can make the product Problem: no champion Alliance Only members of alliance can use Problem: holding alliance together

Control (V&S) Control standard and go it alone If several try this strategy, may lead to standards wars

Openness v. Control Best Strategies: Have a large share of a small market (small pie), Have a small share of a large market (Sun and Java). The optimal point is where you maximize Total Value to you. Systems - are a cooperation of compoenets To caputre value from one component requires cooperation with those providing the other components. Suppliers will want a share of the rewards in retrurn for cooperation.

Openness Strategy When no one firm is strong enough to dictate technology standards. When coordination is necessary. To start the bandwagon rolling. Full openness (ITU standards) v. Alliance Strategy (member advantage, Microsoft developers) Alliance types: special interest group, task forces, groups of companies. Alliances share: cross license patents, manufacturing skills, improve time to market, share designs. Alliances coordinate: standards, interfaces, protocols, specifications.

Control Strategy Central actor controls development Sun sponsors Java, Apple sponsors Macintosh, Ericson sponsors BlueTooth. Companies that control product standards and interfaces have power. But they can lose with poorly conceived standards.

Generic Strategies for new Technology Introduction. 1. Performance Play - new, incompatible, strong control, high risk, big technology advantage, new entrants try with no base to worry about. 2. Controlled Migration - give up performance to reduce customer switching costs, need allies, backward compatible, but proprietary.

Generic Strategies for new Technology Introduction. 3. Open Migration - new product by many vendors, low switching costs, good where there are only manufacturing economies, want a big network. 4. Discontinuity - need allies, make system very open, new product not compatible with old but available from many suppliers, favor efficient manufacturers.

Performance Play (V&S) Introduce new, incompatible technology Examples Palm Pilot Iomega Zip Attractive if Great technology Outsider with no installed base

Controlled Migration (V&S) Compatible, but proprietary Examples Windows 98 Pentium Upgrades

Open Migration (V&S) Many vendors, compatible technology Examples Fax machines Some modems

Discontinuity (V&S) Many vendors, new technology Examples CD audio 3 1/2” disks

Summary of the lecture (V&S) Positive feedback means strong get stronger and weak get weaker Consumers value size of network Works for large networks, against small ones Consumer expectations are critical Fundamental tradeoff: performance and compatibility

Summary on introducing services and products into markets (V&S) Fundamental tradeoff: openness and control Generic strategies Performance play Controlled Migration Open Migration Discontinuity Look at lessons of history