Economics of Standards in Information NetworksStandards play a prominent role in systems characterized by interaction. In information systems, standards provide for compatibility and are a prerequisite for collaboration benefits. More generally speaking, standards constitute networks. In this work, a standardization framework based on an analysis of deficiencies of network effect theory and a game theoretic network equilibrium analysis is developed. Fundamental determinants of diffusion processes in networks (e.g. network topology, agent size, installed base) are identified and incorporated into a computer-based simulation model. As a result, typical network behaviour (specific diffusion patterns) can be explained and many findings from traditional network effect theory can be described as special cases of the model at particular parameter constellations (e.g. low price, high density). On this basis, solution strategies for standardization problems are developed, and a methodological path towards a unified theory of networks is proposed. |
Contents
1 Introduction | 1 |
11 Standards in Information Systems | 2 |
12 Motivation and research questions | 3 |
13 Structure of the thesis | 5 |
14 Methodology and definitions | 7 |
143 Empirical data from Fortune 1000 study | 9 |
2 Standardization problems | 11 |
212 From production side to demand side economies of scale | 12 |
4327 Summary | 132 |
434 Individual consequences | 135 |
435 Agent size | 138 |
436 Variations of network structure | 139 |
437 Installed base effects | 144 |
44 A different decentralized anticipation calculus | 151 |
45 Adapting the diffusion model | 157 |
4512 An integration | 159 |
22 Network effect theory as theoretical foundation 221 Basics | 14 |
2211 General findings of network effect theory | 15 |
2212 Related literature | 17 |
222 Path dependency lockin and multiple equilibria | 19 |
223 Instability tippy networks | 24 |
The startup problem | 25 |
Too much standardization | 26 |
227 Paretoinferior market results and monopolies | 27 |
228 Laissezfaire vs dirigisme | 28 |
23 Reconsidering network effect theory | 29 |
231 Network effects versus network externalities | 32 |
232 Direct vs indirect network effects | 33 |
233 Inefficiency and market failure | 35 |
234 Costs of network size | 37 |
235 Homogeneous network effects | 38 |
236 Confusion of centralized and decentralized decision making | 39 |
beta isn t always better | 42 |
2373 Nuclear power reactors | 43 |
Gas vs Gasoline | 45 |
238 Normative Implications | 46 |
239 Summary | 47 |
3 A standardization framework | 49 |
311 A standardization model for centrally coordinated networks | 52 |
312 A standardization model for decentrally coordinated networks | 54 |
32 Equilibria in standardization problems | 57 |
321 Information | 58 |
323 Equilibria | 59 |
3242 KaldorHicks efficiency | 60 |
33 Standardization games 331 Standard standardization games | 61 |
332 Static noncooperative 2player standardization games | 64 |
333 Sequential noncooperative 2pIayer standardization games | 67 |
3332 Complete imperfect information | 70 |
3334 BayesNashEquilibria | 72 |
334 Dynamic nplayer standardization games | 75 |
3341 Incomplete imperfect information | 76 |
4 A simulation model for the standardization problem | 80 |
412 Simulation parameters and pattern | 82 |
42 The basic standardization problem | 85 |
42 12 Stability of the standardization gap | 88 |
422 The multiperiod basic standardization problem | 91 |
4222 Single standard implementation vs continuous license costs | 94 |
423 Sequential choice | 96 |
424 Individual consequences | 98 |
4242 Wrong decisions | 100 |
425 Agent size and standardization time | 102 |
4252 Stationary state period and network size | 105 |
4261 Errors in different topologies | 109 |
4262 Standardization costs and network density | 110 |
4263 Summary | 112 |
427 Installed base effects | 113 |
428 Summary of multiperiod one standard problems | 117 |
431 The standardization gap and the equilibrium process | 119 |
4312 An adapted Herfindahl index | 121 |
4313 Simultaneous choice with reversible decisions | 124 |
432 Five diffusion patterns | 127 |
4324 Oligopoly | 128 |
4513 Further research | 162 |
5 Solution mechanisms and empirical data | 164 |
51 Empirical data for the standardization problem | 165 |
512 The importance of network effects | 167 |
513 Diffusion of EDIStandards | 168 |
costs and benefits | 171 |
5141 Empirical EDI benefits in the literature | 172 |
5142 Empirical EDI costs in the literature | 173 |
515 EDI data from large enterprises | 174 |
516 EDI data from SMEs | 176 |
5161 Results of the retailers survey | 177 |
5162 Results from the producers and wholesalers survey | 180 |
517 MIS managers opinion on general standardization issues | 183 |
52 Closing the standardization gap | 184 |
522 Problems of centralized coordination | 185 |
5221 Data problem | 186 |
5223 Implementation problem | 188 |
53 The implementation problem | 191 |
531 A case of X500 Directory Services | 192 |
5311 Directory Services | 193 |
5312 Empirical data | 194 |
5313 Profitability analysis | 196 |
532 A network KOI | 203 |
5322 An nplayer solution | 206 |
533 A network ROI for a virtual principal | 207 |
534 Problems associated with the ROI | 209 |
54 The data problem | 210 |
541 An adapted Groves mechanism for the decentralized standardization model | 211 |
5412 A Groves example Let K 1000 and n3 | 212 |
5413 A Groves mechanism for decentralized standardization problems | 213 |
5414 Example of dominant strategies for the adapted Groves mechanism | 214 |
5415 Summary | 220 |
55 A bidding mechanism | 221 |
551 Two examples | 223 |
552 A comparison between the bidding mechanism and centralized coordination | 226 |
56 Managerial and policy implications | 229 |
561 Some general findings | 230 |
562 Networks as a competitive advantage | 231 |
563 Committees and consortia | 232 |
564 Asymmetric network costs and gains | 233 |
565 Infrastructure subsidies and SME integration | 234 |
towards an interdisciplinary network theory | 236 |
611 Absence of externalities | 237 |
613 Exclusion principle | 238 |
615 Separation of consumers and producers | 239 |
62 Towards an interdisciplinary theory of network effects | 240 |
63 Required modeling power of an interdisciplinary theory of network effects | 243 |
633 Emergence of system components and links | 244 |
7 Conclusions and further research | 245 |
72 Further research | 248 |
Variables and symbols | 253 |
List of equations | 257 |
List of figures | 260 |
List of tables | 265 |
268 | |