Breakthough Ideas (continued, with some additions)
The Mobile Telecommunication Standards War
What Distinguishes Winners in a Standards War? According to Shapiro and Varian (1999; pp. 270-276), past winners in network markets possessed seven key assets: (1) Control over an installed base of customers; (2) Intellectual property rights; (3) Ability to Innovate; (4) First-mover advantages; (5) Manufacturing Abilities; (6) Strength in complements; and (7) Reputation and brand name. Moreover, winners skillfully used two basic tactics: (1) preemption¾build an early lead, so positive feedback helps you and hinders your rival; and (2) expectations management¾assemble allies, make grand claims for your product, and knock down the viability of any emerging new technology.
Shapiro and Varian (p. 263) would classify the 2G standards war as Rival Revolutions, two incompatible 2G technologies that also were incompatible with 1G’s FDMA. It pitted AT&T’s TDMA and Motorola’s iDEN in America, and Ericsson and Nokia’s GSM, in Europe and international markets, against upstart Qualcomm’s CDMA. Founded in 1985, late-comer Qualcomm did not introduce the concept of CDMA until November 1988, demonstrate it until November 1989, secure cdmaOne standardization as IS-95 until 1993, or establish its first commercial network in Hong Kong until 1995, four full years after GSM in Sweden. Qualcomm had no installed base; its rivals dominated the installed base for radio, wire line, analog, and digital wireless markets. Qualcomm was a late-mover; its rivals were first-movers with greater first- and second-generation learning and experience. Qualcomm had no value chain; its rivals possessed greater manufacturing competence and had created scale economics. Qualcomm’s had no complements; its rivals had services, like SMS, that complemented their voice product. Qualcomm was a start-up; its rivals had established international reputations and strong brands. Qualcomm’s assets were limited to its intellectual property rights and ability to innovate; but its rivals challenged their claims to both.
Tactically, their rivals had preempted Qualcomm by already igniting their own positive feedback. Their rivals also proved to be masters of expectations’ management. When Jacobs touted the technical advantages of CDMA, his competitors labeled him “mendacious” and his claims “exaggerated, even impossible,” as mere “Jacob’s Patter,” as his “Blind Faith” that “violated the laws of physics.” (Time has proven who speaks truth and who has forked tongue.)
However, as Shapiro and Varian (p.197) also recognized, “Strength in a network market is measured along three primary dimensions: existing market position, technical capabilities, and control of intellectual property such as patents and copyrights.” Qualcomm had the latter two. Never underestimate the power of breakthrough ideas to create wealth.
The 3G War Begins. Believing a global 3G platform could attract two billion users by 2010, the ITU sought to encourage 3G standards that would harmonize frequency spectrum and air interface standards worldwide through its IMT-2000 initiative. The ITU specified that 3G technologies should achieve packet-based transmission of digitized voice, data, or video and permit global roaming. Its guidelines called for peak data speeds were 144 kbps for mobile and 2 Mbps for fixed.
In his chapter on The Genesis of a Single Flexible Standard, Daniel Steinbock (2001, p.210) captured the essence of Europe’s coalition in the 3G standards war, “To paraphrase Carl Von Clausewitz, the function of these alliances has been competition [Clausewitz’s war] through other means.” He described the EU as seeing the 3G competition as the most important opportunity for European companies to reverse strategic positions with Silicon Valley and consumer electronic giants in Japan. Their primary objective was to ensure the compatibility of the 3G standards with GSM. Seeking to circumvent Qualcomm’s patented intellectual property, Nokia and Ericsson led its European coalition toward a standard that attempted to homogenize (bastardize?) Qualcomm’s upstream 3G innovations.
In 1998, agreeing with the ITU’s call for unity, Qualcomm argued that 3G needed a single standard, but also, through their proposed multiple standards, that the EU was intent on locking American technology out of the European market. The Clinton administration examined this trade conflict. However, Nokia successfully lobbied Washington and the U. S. press. In August 1998, Business Week’s cover story portrayed Nokia as the “way of the future.” (Only Americans know what a Business Week cover really means--that you have topped out.) At the same time, seeking to ignite a worldwide network effect, ETSI announced a preliminary decision requiring its members to use its 3G standard, just as it had mandated GSM as Europe’s 2G standard.
In a letter to the ITU, Qualcomm denounced this required use of the ETSI standard as old-style protectionism and threatened to use its patent portfolio to protect its intellectual property. In mid-December, the EU lent its support to the ETSI decision. In response, a letter of protest, signed by the U.S. Secretary of State, Trade Representative, Secretary of Commerce, and FCC Commissioner, was sent to the EC, the EU Commissioner. Also in December 1998, Finland began to issue its 3G licenses, giving its imprimatur to this fait accompli. Thus, Europe sought to parlay its GSM strategy and installed base to dominate 3G mobile telecommunications.
In January 1999, the EU Commissioner responded that its intent was not to lock out U. S. players, but only to ensure that at least one operator in each European country would use compatible technology so that European phones would work all across Europe. These “good intentions” seemed to appease the U.S. officials, demonstrating the power of Euro-spin. The U.S. did send an unsigned “verbal note” to Finland’s Ministry of Transport and Communication objecting to its beginning licensing WCDMA before patent disputes were settled.
The spin from Finland/Nokia was that they simply favored “technology agnosticism.” Steinbock (p. 224, his emphasis) reported the spin-as-gospel, “The point had not been how a given technology works, but what it could do for the customer.” Qualcomm believed that how a technology was designed and worked precisely determined what it could do for customers.
The Decisive 3G Battle. At the same time, a major legal suit with Ericsson was ongoing. Beginning in December 1995, after claiming repeatedly for years that CDMA could never work, Ericsson changed its tactics from “it cannot possibly work” to “we invented it here first” by announcing it held eight essential blocking patents for IS-95. Given how few GSM patents (only four obtained by acquisition) Ericsson held in 1995, this claim must be characterized as surprising to say the least. In September 1996, just as the first commercial 2G networks were rolling out in South Korea and the U.S., Ericsson sued Qualcomm for patent infringement. Qualcomm counterclaimed unfair competition against Ericsson.
On October 13 1998, during the culminative battles of the 3G standard’s setting war, Qualcomm supported a single converged worldwide standard that met a set of technical criteria and fairness principles. It proposed resolving disputes on technological points by selecting the proposal that was demonstrably superior in terms of performance, features, or cost. If there were no material differences, then the choice that was most compatible with existing technology was to be favored. In a letter to the ITU, moreover, Qualcomm threatened to withhold licensing of its intellectual property, which was essential for the proposed WCDMA standard.
One week later, on October 20, 1998, Qualcomm announced that Ericsson had dismissed it claims under three patents asserted against Qualcomm; and, in a separate transaction with the U.S. Patent and Trademark Office, Ericsson admitted the invalidity of two more of its eight “blocking” patents. Unlike Qualcomm, whose patents on IS-95 had already been licensed by 55 companies, no company had licensed the “eight essential” patents from Ericsson. Qualcomm pointed out that these merit-less claims revealed Ericsson’s anticompetitive motives in falsely claiming essential patents.
This IP battle was the prelude to Ericsson’s surrender. On March 25 1999, Ericsson and Qualcomm reached a global CDMA resolution that supported three worldwide standards. They cross-licensed their respective patent portfolios, but only Qualcomm received royalties. In addition, Ericsson purchased Qualcomm’s wireless infrastructure business. Thus, Qualcomm emerged with a tremendous victory that: (a) announced a pragmatic European endorsement of Qualcomm’s intellectual property position in the standards war; (b) paved the way for expansion of the global CDMA market, because each standard had spread spectrum architecture as its core; (c) removed Qualcomm from the expensive and low-margin infrastructure business¾except for being the world’s only 2G CDMA CSM supplier; (d) secured Ericsson’s membership in Qualcomm’s value web and its additional royalty income; and (e) promulgated the significant technological complexity of three variations of CDMA-3G-standards, a complexity which favored Qualcomm’s network asset of a superior ability to innovate¾including the ability to solve the problems within Europe’s preferred, but technologically-bastardized, WCDMA standard. Ericsson succeeded in getting a head start in learning how to install CDMA infrastructure, and more significant strategically, it had hedged its bet on WCDMA.
If you look beneath the cool business prose, it was a stunning victory for Qualcomm. Because data services were becoming an inevitable and major part of the wireless future, not only had Europe been forced to accept spread spectrum architecture for 3G, but also Ericsson chose to consort with the enemy. Ericsson, who claimed Qualcomm had invented a CDMA system that would never work, who next claimed to have invented it first, had surrendered and joined Qualcomm’s value web. Why would Ericsson hedge their bet on WCDMA and seek to improve their CDMA spread spectrum skills? Because Qualcomm was shaping the industry’s future by the power of its integrated learning base in spread spectrum technology.
The WCDMA initiative, which is “modestly” called the Universal Mobile Telecommunication System (UMTS), failed in its attempt to bypass Qualcomm’s intellectual property through technical “improvements” supposedly drawn from its GSM core competencies. Instead, its selected 3G standard merely restricted the future flexibility of European carriers, which was bad enough, but still worse, it introduced inherent technical problems into its global design rules for certain parameters within its version of spread spectrum.
Specifically, problems associated with asynchronicity, higher than ideal chipping rates, and narrow guard bands deviated from the theory of spread spectrum modulation or were not based on basic empirical tests that might prove their value as an integrated system. How could Europe possibly design an ideal modular UMTS system when it had no experience with the complex interdependencies in spread spectrum modulation? How could it master the engineering complexity of advanced power control for a spread spectrum network when it had no experience with power control or spread spectrum? How could it add advanced features when the complexity of managing the interdependencies in spread spectrum was, as an engineering practice, still terra incognita?
Attempting to bypass Qualcomm’s IPR proved impossible anyway because certain key solutions to basic problems, like the waveform and methods for spreading the spectrum, proved to be not only ideal but also essential for any workable spread spectrum system. That is why every member of the GSM Grand Alliance has paid to license Qualcomm’s IP and contractged pay royalties. Whereas, Qualcomm received cross-licensing right to GSM IP for its world phones, but without an paying royalty. Can you not understand the implication of these economic transactions? Not only was Europe’s course a political and strategic failure, but also this disastrous political process of designing without having the necessary practical experience in the spread spectrum RAN, the engineering experience and tacit knowledge which permits and builds a comprehensive understanding of the scope of interactions and how to shield the system from endless patches required to resolve these reciprocal interdependencies that continually effect another part of the unstable new system, will prove even more costly than all of their expenditures of time, money, and effort at patchwork solutions because it cost Europe even more dearly in the battle for the future.
Its supposed UMTS migration path was a fantasy of an assured pathway to advanced services when sold at auction, but proved a disaster at the time of implementation (if the delays ever end). These unnecessary complexities in their chosen global design rules are proving to be serious, but still profitable, problems that only the spread spectrum technology leader, Qualcomm, could best solve.
Worse yet, the newly auctioned spectrum for its two 5 MHz channels, mandated by the design rule for separate forward and reverse channels in different and exclusively reserved bands of spectrum, cost the operators a fortune, $100 billion euros. Moreover, Europe once again required at least one selected standard needed to be cross-European to ensure global roaming and that operators must declare which standard they would use at auction, locking them all into a bad decision.
What’s more, the European operators must have actually believed that the GSM MAP compatibility somehow ensured a RAN compatibility that promised an easy migration and low adoption costs. The operators were used to believing in the technical claims of their infrastructure providers because everyone knew that Europe was master of wireless telephony, and the Musketeer’s Oath meant that all Europeans were for Europe.
Given European expectations and beliefs, no operator would have dared choose Qualcomm’s CDMA as a standard. It was an option of multiple possibilities but only one real choice. Once again Europe bet on their repackaged 2G GSM strategy for UMTS that had once gained them world market hegemony. But this time, they knew not what their coalition of centralized political planning had wrought. What remarkable handicaps to inflict on European operator’s costs and margins!
Once again, the key point for investors to understand is that the 3G standards war is the type that Shapiro and Varian classify as Evolution versus Revolution. That is, Qualcomm’s advanced performance in CDMA2000 1x remains evolutionarily compatible with IS-95 because both use identical spread spectrum architecture. The global design rules remained intact; the basic interfaces unchanged, and the tools and procedures for testing integration of the RAN basically similar. Because of its failed efforts to circumvent Qualcomm’s IPR, however, UMTS demonstrates no compelling performance advance to make up for its high adoption costs (which are RAN switching costs by another name). UMTS is merely an inferior and problematic variant of the ideal spread spectrum performance that was designed and commercialized by Qualcomm.
Although third-generation UMTS must be classified as a “revolution” in the context of a standards war because its architecture’s global design rules, parameters, interfaces, and integration tools were discontinuous and incompatible when compared to second-generation GSM, UMTS does not even advance, much less leapfrog, Qualcomm’s continuation of its normal trajectories that advance CDMA performance.
In a two-way battle for architectural control of an emergent de facto standard for 3G spread spectrum, Qualcomm exclusively owns the intellectual property and follows the evolutionary path leading from 2G cdmaOne to 3G CDMA2000. But also, Qualcomm owns the essential IPR for UMTS, and European vendors own bits and pieces of the less essential IPRs for its European-identified UMTS standard. Anyone pursuing the path to 3G UMTS from 2G GSM must take an architecturally incompatible path to get there, and moreover, one with much less potential intrinsic value in projected cost/performance ratio than CDMA2000.
UMTS necessarily remains architecturally incompatible with GSM’s air interface. Vastly different global design rules translate into a generational discontinuity, necessitating higher adoption costs. What is worse, the path to UMTS that runs through GPRS and EDGE keeps the European vendors and operators locked into more of the same old and tired TDMA technology because these stepping stones are mere extensions of the GSM global design rules.
When the game changed to spread spectrum, universal Darwinism says, “adapt or die.” In January 1998, ETSI endorsed spread spectrum as its next generation standard to replace GSM. From that point on, European vendors and operators should have begun to become experts in spread spectrum architecture and operation. GPRS or EDGE intermediaries waste more time, money, intelligence, and effort once it was clear that the third-generation technology game was to be played using global design rules coming from a spread spectrum architecture. These are disastrous design and business choices in an age where architectural control is so strategically valuable. How can this significant distinction still be below many, maybe most, operators,’ vendors,’ analysts,’ and investor’s radar? Who says ideas do not create wealth?
So, who owns the seven strategic assets in the 3G Standards War? Recall that past winners in network markets possessed seven key assets: (1) Control over an installed base of customers; Europe and most analysts still believe UMTS somehow controls the GSM installed base of customers, but the switching costs for customers are negligible, the cost of a phone and paying or waiting out an old contract, and switching costs are continually being reduced for carriers by overlay and world phones. (2) Intellectual property rights; Qualcomm owns the most significant IPR, and UMTS promises to increase royalties to some as yet unknown level as fair compensation for their “inventions,” significant advantage Qualcomm, who pays for the next generation from its UMTS royalties and drives up the costs of operators who choose the UMTS alternative; (3) Ability to Innovate; Qualcomm ahead by three breakthrough lengths; (4) First-mover advantages; Qualcomm so far in the lead that the ratio of adoption for 3G subs and handset replacements surely must be misleading, as if a three-year-lead in 3G commercialization were not a significant and sought after competitive advantage; (5) Manufacturing Abilities; Asia coming on fast on the inside track as masters of color LCDs, miniaturization, and consumer electronics within CDMA terminals;(6) Strength in complements; Qualcomm’s position location, BREW, and soon QChat are trump cards for Qualcomm, KDDI, KTF, and China Unicom; and, (7) Reputation and brand name; Nokia still in the lead but starting to fade, with Samsung gaining ground, but still worried about China (but not Nokia who has given up in Korea); with Qualcomm still despised, just like a gorilla always is. |
