1. 1 Prentice Hall, 2002 (for lecture 21.01.03) Information Rules - Shapiro & Varian Chapter 1: The Information Economy Chapter 7: Network Economics modifications by J.Molka-Danielsen

2. 2 Prentice Hall, 2002 The Information Economy Carl Shapiro Hal R. Varian

3. 3 Prentice Hall, 2002 Systems of Products Complementary products Hardware/software Client/server Viewer/content Product lines High fixed cost, low incremental cost Leads to value based pricing

4. 4 Prentice Hall, 2002 Unique Features Complements Different manufacturers Strategy for complementors as well as competitors Compatibility as strategic choice Standards and interconnection Product lines Lower quality may be more expensive

5. 5 Prentice Hall, 2002 Information Anything that can be digitized Text, images, videos, music, etc. a.k.a. content, digital goods Unique cost characteristics Unique demand characteristics

6. 6 Prentice Hall, 2002 Cost structure Expensive to produce, cheap to reproduce High fixed cost, low marginal cost Not only fixed, but sunk No significant capacity constraints Particular market structures Monopoly Cost leadership Product differentiation (versioning)

7. 7 Prentice Hall, 2002 Rights Management Low reproduction cost is two-edged sword Cheap for owners (high profit margin) But also cheap for copiers Maximize value of IP, not protection Examples Library industry Video industry

8. 8 Prentice Hall, 2002 Consumption Characteristics Experience good Browsing Always new Reputation and brand identity Overload Economics of attention Hotmail example Broadcast, point-to-point, hybrid

9. 9 Prentice Hall, 2002 Technology Infrastructure to store, retrieve, filter, manipulate, view, transmit, and receive information Adds value to information Web = 1 terabyte of text = 1 million books If 10% useful = 1 Borders Bookstore Value of Web is in ease of access Front end to databases, etc. Currency

10. 10 Prentice Hall, 2002 Systems Competition Microsoft-Intel: Wintel Intel Commoditize complementory chips Microsoft Commoditize PCs Apple Integrated solution Worked better, but lack of competition and scale led to current problems

11. 11 Prentice Hall, 2002 Lock-In and Switching Costs Example: Stereos and LPs Costly switch to CDs Systems lock-in: durable complements Hardware, software, and wetware Individual, organizational, and societal

12. 12 Prentice Hall, 2002 Network Effects Value depends on number of users Positive feedback Fax (patented in 1843) Internet (1980s) Indirect network effects Software Expectations management Competitive pre-announcements

13. 13 Prentice Hall, 2002 Compatibility Examples Beta v. VHS Sony v. Philips for DVD Role of 3rd parties Read v. write standards Backwards compatibility? Windows 95 Windows NT

14. 14 Prentice Hall, 2002 Basic Strategies Go it alone Partnerships (Java) Formal standard setting Widespread use Licensing requirements Competition in a market or for a market?

15. 15 Prentice Hall, 2002 Policy Understand environment IP policy Competition policy Regulation Antitrust Electronic commerce Contracts Privacy

16. 16 Prentice Hall, 2002 Information is Different… but not so different Key concepts Versioning Lock-in Systems competition, Network effects

17. 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

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

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

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

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

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

23. Positive Feedback n Positive feedback - the strong get stronger and the weak get weaker. n Speaking into a microphone, when the amplified signals go back into the system, you have repeated amplification. It gets loud. n Negative feedback - when the strong get weaker and the weak get stronger. Go to a middle point. Industrial oligopolies exhibit. n 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.

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

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

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

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

28. S-shaped curve n 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. n Seen in the adoption of: fax machine, CD, color TV, video games, e-mail, and the Internet.

29. Economies of Scale n 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. n In manufacturing the benefits of scale are usually exhausted before total market dominance. n 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.

30. Network externalities n 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) n Recall externalities are positive and negative (pollution) n Positive network externalities lead to positive feedback n 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) = n 2 -n where n is the number of users. If value=$1 for a single user, and n=10, then network size 10 has 10 2 -10=90 = $90.

31. Collective switching costs n 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. n 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. n 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.)

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

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

34. Industry and positive feedback n 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.) n 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.) n 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.

35. 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.

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

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

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

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

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

41. Openness v. Control n Is the second fundamental tradeoff. n Open - make interfaces and specifications available to others. Consumers can fear lock-in, open raises chances of success, attracks allies. n 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. n Existing market position, technical capabilities and control of intellectual property rights are 3 dimensions of critical market strength. n 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.

42. Openness v. Control (V&S) n Your reward (total value to you) = Total value added to industry x your share (in the market). n 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). n 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).

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

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

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

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

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

48. 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.

49. 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.

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

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

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

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

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

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