9.14 – Software litigation: a feast for the lawyers
Source: Montage – sources in art itself.
By 2000 the personal computer industry had become one where high margins were earned only by those with a stake in the branded components of the PC. By and large the PC itself is not a branded product. It was Intel and Microsoft, the dominant suppliers of chips and software respectively, who had come closest to monopoly positions, with large market shares in their market segments. It is difficult to provide direct analogies with other industries, since most products are differentiated by what they do, how they work, how they look, how much they cost and what after-sales service they provide. External styling had rarely been a major differentiating factor for PCs, and after-sales service diminished in importance when reliability levels increased and PCs became increasingly homogeneous with many common components. The producers of personal computers have therefore had to compete increasingly on price alone. This changed the whole structure of the industry, as it meant that growth on its own did not necessarily produce high profit margins. This had nothing to do with globalisation and the new economy, and everything to do with the development of Intel and Microsoft as industry ‘equalisers’.
9.15 – Dell vs Compaq: only one winner
Source: Compaq and Dell annual reports. CRSP, Center for Research in Security Prices, Graduate School of Business, University of Chicago, 2000. (Used with permission. All rights reserved. www.crsp.uchicago.edu.)
The impact on industry structure can be seen in two ways. Firstly, just as Compaq provided a lower-cost alternative to IBM, so it too came under threat later from another low-cost producer. The difference was that Dell was not simply a low-cost producer, but also a low-cost supplier, having extended the logic of low cost from the production chain into the distribution channel. Dell emerged just as mail order reached new heights. In a world where the supply chains could be constructed in electronic form and integrated with production, it was suddenly possible to remove large elements of cost from the system. This took the form not only of reduced administration and bureaucracy, but also much lower working capital requirements (work in progress and inventory). Since PC production was largely an assembly process, if the components could be moved onto a ‘just in time’ basis there would no longer be the same levels of capital and cost involved in holding components. This change in the industry, and the importance of the cost curve, can be seen in the way in which Dell overtook Compaq (figure 9.15). Dell fully exploited the importance of being the low-cost producer, but was unable to escape the accompanying logic. Irrespective of how efficient it became, it remained in a low-margin, high-volume business, which was therefore sensitive to the whims of the economic cycle. Growth might remain strong, but would be cyclical; the margins would be retained by those with proprietary products.
The PC business in perspective
The assembly-driven structure of the personal computer industry was very similar to the early days of the automobile industry, when producers were effectively assemblers of externally produced components and financed purchases and production through credit and customer advances. The automobile industry transformed itself into a volume business when the automobile became a mass-market product and economic downturn culled most of the competition. The PC business was a volume-driven one almost from the start. Brand brand loyalty has been progressively removed by standardisation of components and software. Minor production or pricing errors can have an immediate and major impact on the income statement. Like the automotive industry, the PC business remains susceptible to an economic downturn. The lack of an inventory buffer means that any economic downturn will have an instantaneous impact on suppliers, and volatility throughout the supply chain will increase. The ‘shock’ element will therefore undoubtedly be much higher than hitherto. If this ‘shock’ were met with large-scale retrenchment, the overall economic impact would be amplified.
While the software industry produced an industry colossus in Microsoft, the hardware industry continued its evolution from one driven by the economics of production to one driven instead by the economics of distribution. The standardisation of the personal computer, and its metamorphosis into an assembled item with ‘named’ components (most notably the microprocessor), meant that reliability became less of an issue. Consequently, the importance of the ‘PC’ brand name diminished and was replaced by the ubiquitous ‘Intel Inside’ as an effective product quality guarantee. As a consequence the principal factor became price, the price at which the PC could be delivered to the consumer.
New companies emerged that had built operating models which minimised the costs associated with sales and after-sales service as well as through efficient production. Those that could move to the lowest point on the cost curve were in a position to make profits, those that could not reduce their costs to this lowest level had to operate on wafer-thin margins or even at a loss if they wished to retain market share. The personal computer industry became a commodity one, identical in most respects to the traditional ‘old’ industries of chemicals or steel. It used to be a ‘growth cyclical’ in the sense that the market was expanding at rates in excess of GDP growth. As a consequence, while it was prone to sharp swings in profitability and inventory requirement, there at least remained the comfort of underlying growth. This produced the apparently paradoxical situation of an industry with both growth characteristics and declining margins. The move to mobile devices has taken away this growth advantage so that PC producers are effectively now ‘old economy’.
The software industry was part of this enabling in the sense that just as ‘Intel Inside’ homogenised the hardware so ‘Windows’ did the same for the software, and between the two this effectively commoditised the PC. Applications have evolved, with some products having been overtaken by the development of their competitors – the most obvious example being VisiCalc being overtaken by Lotus 1-2-3, which in turn was overtaken by Microsoft Office – but the almost universal adoption of Windows created a form of monopoly protection. This was very similar to the barriers to entry that protected Bell, Edison and RCA.
When VisiCalc was first produced, the application of the laws of patent and copyright to software was not obvious. However, it did not take long for developers to realise the need to protect their products if high margins were to be sustained. Equally, Apple, which had seized upon the work originated by Engelbart and others to produce a generational shift in the visual display and ease of use of programs, was unable to monopolise this work in the way that Microsoft was able to for the IBM-clone-based market.
Unlike the Apple market, which remained a relatively niche part of the overall personal computer sector, the IBM clone segment was the volume arena and Microsoft the provider of the operating, interface and applications software, with the former providing a huge benefit in the delivery of the latter. The strength of this position placed Microsoft in a monopolistic position in much the same manner that AT&T controlled the early development of the telephone industry or Standard Oil the oil industry. The level of power this conferred brought an identical response from the competition authorities regarding potential abuse and predatory actions. Microsoft adopted a combative approach to the intrusion of regulators, although at various stages it was forced to allow access to third-party software. Nevertheless, history would have suggested that in the absence of meaningful concessions the end result of this would have been some form of enforced breakup. However, market forces can be more powerful than those of the regulator. The dominance of Microsoft’s position in operating systems was overtaken by events. The threat of the Internet was one which the company, bound by the legacy of its installed base, was not well-equipped to deal with. The history of the business had been in PCs; the future was in mobile devices. One might have expected that Microsoft had a competitive advantage because of customer familiarity with its operating system but in some ways this ended up being a disadvantage as new dedicated operating systems were created by competitors
. In launching the iPhone Apple created a mobile interface (iPhone OS, later rebranded iOS) that was intuitive and simple to use, and opened the door to the creation of a family of products linked through its closed ecosystem, which included iTunes and the Apps Store. Google sought to protect its search revenues by purchasing and further developing a mobile operating system (Android) that embedded the use of its search engine, which it complemented with its own apps ecosystem and maps services.
Technological evolution has made the monopoly question much less pressing for Microsoft but the bull’s eye that was on its chest has moved to new targets. It is worth remembering that Standard Oil’s response was to fall back on legal argument, resist the overtures of government and seek to counter the persistent attacks in state and federal courts. The eventual outcome was the breakup of the trust into its constituent companies. For the investor, this outcome was not a damaging one, since the individual entities were sufficiently large in size to become profitable and growing companies in their own right. As a consequence, it did not prove a financial penalty for Rockefeller either. What was decisive to the profitability of the industry was its control over distribution and pricing. This was partially removed by the breakup, but more than compensated for by the fact that the breakup occurred just as the automobile industry was moving into a period of rapid growth.
9.16 – Success brings problems: Microsoft monopoly investigations
Source: Montage – sources in art itself.
So far as AT&T was concerned, in the J. P. Morgan/Theodore Vail era, the approach adopted by government competition concerns was conciliatory. AT&T acquiesced to a degree of federal control, and as a consequence managed to retain the local and long-distance networks and the virtual monopoly that these conferred. Part of the penalty for this was that the company had to be very careful about what it did in related fields. Many potential growth opportunities that came out of the research in Bell Laboratories were left to be exploited by others. While Microsoft no longer retains the same level of monopoly control despite the persistence of Windows and Office, it is still worth remembering the degree of animosity felt towards the company during the late 1990s.
Summarising the structure of the personal computer industry at the turn of the century is probably most easily achieved by reference to the margins earned by the different players. Figure 9.17 clearly shows the dominance of Microsoft and Intel, even by comparison with Dell, the most successful new entrant. On the hardware side, it was Intel microprocessors that largely defined the hardware. On the software side, it was the Microsoft family of products which the consumer identified.
Dominance of this magnitude could not continue indefinitely. At the time Intel had had competitors such as Advanced Micro Devices (AMD), whose products frequently rivalled those of Intel. However, AMD’s challenges proved to be short-lived if persistent. Barriers to entry had risen since the early years, though, when intellectual and technical capability were the dominant characteristics. Such is the sophistication and precision of the end product that these earlier characteristics have been overlaid with the demands of process technology, yield and capital. A single production facility would now cost in excess of $1bn to build, which presents a formidable barrier to most potential entrants, who would not only need to find such capital but also sustain volume production against entrenched players with heavy setup costs. The shift to mobile proved difficult for Microsoft, as it also did for Intel. Despite repeated attempts Intel found it had ceded leadership to ARM in mobile chips design in a world where designers designed and fabricators produced.
When compared with other technological advances, the personal computer industry can be seen as an offshoot from the mainstream development work largely funded by government defence requirements. This development work produced technologies that were picked up by groups of individuals who sought to transform them into a viable commercial product. From that point the traditional cycle of innovation, propaganda, retrenchment, failure and ultimate success was rejoined. Many companies had periods of temporary success, but ultimately only a small number of players emerged as long-term success stories. As in previous technological waves, the common feature of those that made it was the ability to build barriers to entry.
9.17 – A question of profit: Intel and Microsoft take the lion’s share
Source: Annual reports of Dell, Compaq, Apple, IBM, Microsoft and Intel.
In the personal computing industry, IBM helped to create the mass market, but the manner in which it did so created an industry which transformed itself into an assembly business driven mainly by cost. Part of this development was the creation of what could be termed ‘integrators’, or perhaps quality control checks, in Intel and Microsoft. Because of their superior technology and/or market knowledge, Intel and Microsoft became part of a select band whose profit margins remained protected by these barriers to entry. The barriers to entry were effective until they were circumvented by a product shift away from desktop to mobile. Both companies failed to manage the transition and lost leadership as a consequence. Their historic business remained reasonably robust but without the growth that had been diverted to the new areas.
The other element that comes clearly out of the PC story is how difficult it is to forecast future developments at times of profound shifts in technological capabilities. These shifts were rarely foreseen, even by those whose scientific work helped to create them. As a result, commercial implications entered rarely into financial analysis and valuation work. Once the initial shifts took place, however, those such as Gates who had the clearest vision and the drive to exploit it were to find the path an extraordinarily profitable one. However, the circle of life continued and another ‘initial shift’ followed which was again exploited by those who had the vision, drive and perhaps good fortune.
* * *
84 C. Cerf and N. S. Navasky, The Experts Speak: The Definitive Compendium of Authoritative Misinformation, New York: Villard, 1998, p.231.
85 Ibid.
86 Ibid.
87 W. Aspray, ‘The Intel 4004 Microprocessor: What Constituted Invention?’, IEEE Annals of the History of Computing, vol. 19, no. 3 , 1997, pp.4–15.
88 Ibid.
89 P. Freiberger and M. Swaine, Fire in the Valley: The Making of the Personal Computer, New York: McGraw-Hill, 2000, p.22.
90 Ibid.
91 B. Rosen, Morgan Stanley Electronics Letter, 11 July 1979.
chapter 10
The Internet
How computing timeshare became a global phenomenon
“We’ve all heard that a million monkeys banging on a million typewriters will eventually reproduce the entire works of Shakespeare. Now, thanks to the Internet, we know that this is not true.”⁹²
Robert Wilensky
“The Internet is a shallow and unreliable electronic repository of dirty pictures, inaccurate rumors, bad spelling and worse grammar, inhabited largely by people with no demonstrable social skills.”⁹³
Chronicle of Higher Education, 11 April 1997
“By 2005 or so, it will become clear that the Internet’s impact on the economy has been no greater than the fax machine’s … ten years from now, the phrase information economy will sound silly.”⁹⁴
Paul Krugman, writing in 1998
The chapter is split into four sections and includes updated material to reflect the changes which have taken place since the first edition of the book was published. Part 1 outlines the origins and early development of the Internet. Part 2 considers its commercialisation through case histories of the key companies that emerged in its early history. Part 3 studies the associated stock market reaction, then the extraordinary bubble that ensued – and its aftermath. The final section looks at the wider impact of the Internet and how it may evolve in the future. This theme is also developed further in the final chapter.
Part I: The lure of computer networking
Something stirs in academia
Even as the first electronic computers we
re being introduced, scientists were already speculating on the potential gains that could be had from linking them together. The two principal groups to show interest were the defence sector and academia. In defence, the realm from which mainframe computing had emerged, the nuclear age extended computing demands beyond shipping and ballistics. Computers soon became an integrated part of the nuclear defence command-and-control system. This required the creation of a network of computers and links between dedicated mainframes in defence establishments and those in the research facilities of defence academics. In the case of academics, demand for access to mainframes extended well beyond those directly involved in government-funded defence work. The difficulty was that mainframe computers were expensive to purchase, install and maintain. They could only be justified as purchases by the collective needs of many academic departments and users. There were thus strategic reasons why mainframe computers should be networked in defence, and economic ones in academia.
At a more futuristic level, many individuals could see the potential for linking computers and the benefits that might accrue as a result. The earliest and most important statement of intent came at the end of World War II from Vannevar Bush, the director of the Office of Scientific Research and Development in the US. In Britain, as we saw in chapter 8, Winston Churchill followed the conventional British approach of maintaining as much secrecy as possible about defence matters, thereby inhibiting the spread of knowledge and expertise gained at establishments such as Bletchley Park during World War II. Bush took the opposite approach. President Franklin D. Roosevelt, at Bush’s instigation, requested a report on the future direction of military and scientific research. Bush, a founder of the American Appliance Company (later to become Raytheon), delivered the report, entitled ‘Science: The Endless Frontier’, to President Roosevelt in July 1945. The report was to define the framework for the development of technology in America in the second half of the 20th century.
Engines That Move Markets (2nd Ed) Page 49