Published online by Cambridge University Press: 13 December 2011
Robert Noyce's career at Fairchild Semiconductor sheds light on several developments that were central to the growth of Silicon Valley and the semiconductor industry: entrepreneurship, technical leadership, and the management of growth in a high-technology company. Noyce served as Fairchild Semiconductors first head of R&D and as its general manager for the six years of the company's most dramatic growth. His technical orientation, personal interest in new technologies, and hands-off management style helped establish a culture at the firm that welcomed innovations in research, process technology, manufacturing, and marketing. As Fairchild Semiconductor grew into a multidivisional mass producer, Noyce's entrepreneurial leadership proved inadequate. Communication breakdowns between divisions, coupled with a series of poor decisions by the parent company, further contributed to the decline of Fairchild Semiconductor.
1 San Francisco Chronicle, 10 Aug. 1991; Peninsula Times Tribune, 10 Aug. 1991; plaque at 844 Charleston, Palo Alto.
2 Fairchild Camera and Instrument Annual Report 1966.
3 Tom Wolfe, “The Tinkerings of Robert Noyce: How the Sun Rose on Silicon Valley,” Esquire, Dec. 1983, 346–74. The article ran next to profiles of other “American Originals,” including Jackie Robinson, John F. Kennedy, Betty Friedan, Walt Disney, Elvis Presley, and the brothers Rockefeller. This heroic image of Noyce persisted throughout his life. See, for example, Otis Port, “Bob Noyce Created Silicon Valley. Can He Save It?” Business Week, 15 Aug. 1988, 76.
5 For a sampling of book-length journalistic and popular historical accounts of the rise of Silicon Valley, see Malone, Michael S., The Big Score: The Billion Dollar Story of Silicon Valley (Garden City, N.Y., 1985Google Scholar); Hanson, Dirk, The New Alchemists: Silicon Valley and the Microelectronics Revolution (Boston, 1982Google Scholar); Rogers, Everett M. and Larsen, Judith K., Silicon Valley Fever: Growth of High-Technology Culture (New York, 1984)Google Scholar.
6 Holbrook, Daniel, “Diversity, Complementarity, and Cooperation: Materials Innovation in the Semiconductor Industry,” in Goldstein, Andrew and Aspray, William, eds., Facets: New Perspectives on the History of Semiconductors (New Brunswick, N.J., 1997), 77Google Scholar.
7 Lecuyer, Christophe, “Making Silicon Valley: Engineering Culture, Innovation, and Industrial Growth, 1930–1970,” unpublished Stanford University diss., Aug. 1999Google Scholar, chs. 3–4; Bassett, Ross Knox, “New Technology, New People, New Organizations: The Rise of the MOS Transistor, 1945–1975,” unpublished Princeton University diss., Jan. 1998Google Scholar. Another interesting piece, written by Gordon Moore, who cofounded both Fairchild Semiconductor and Intel (with Noyce) is “The Role of Fairchild in Silicon Technology in the Early Days of ‘Silicon Valley,’” Proceedings of the IEEE 86 (Jan. 1998): 53–62.CrossRefGoogle Scholar
8 Leslie, Stuart W., The Cold War and American Science: The Military-Industrial-Academic Complex at MIT and Stanford (New York, 1993Google Scholar); “How the West Was Won: The Military and the Making of Silicon Valley,” in Aspray, William, ed., Technological Competitiveness: Technological and Historical Perspectives on the Electrical, Electronics, and Computer Industries (Piscataway, N.J., 1993Google Scholar); Saxenian, AnnaLee, Regional Advantage: Culture and Innovation in Silicon Valley and Route 128 (Cambridge, Mass., 1994Google Scholar); and “The Genesis of Silicon Valley,” in Hall, Peter and Markusen, Ann, eds., Silicon Landscapes (Boston, 1985), 20–34Google Scholar. This work is part of a much larger body of research on industrial districts. For an excellent study of the interactions among specialized suppliers and their customers in an industrial district, see Scranton, Philip, Figured Tapestry: Production, Markets, and Power in Philadelphia Textiles, 1885–1941 (New York, 1989Google Scholar).
9 Lee, Chong-Moon, Miller, William F., Hancock, Marguerite Gong, and Rowen, Henry S., “The Silicon Valley Habitat,” in Lee, Chong-Moon et al. , eds., The Silicon Valley Edge: A Habitat for Innovation and Entrepreneurship (Stanford, Calif, 2000), 3Google Scholar. For information on distinctive features of the semiconductor industry, see Langlois, Richard N. and Steinmueller, W. Edward, “The Evolution of Competitive Advantage in the Worldwide Semiconductor Industry,” in Mowery, David C. and Nelson, Richard R., eds., Sources of Industrial Leadership: Studies of Seven Industries (Cambridge, U.K., 1999), 21Google Scholar. Among the most useful general examinations of the early semiconductor industry are Braun, Ernest and MacDonald, Stuart, Revolution in Miniature: The History and Impact of Semiconductor Electronics (Cambridge, U.K., 1978Google Scholar); Levin, Richard, “The Semiconductor Industry,” in Nelson, Richard R., ed., Government and Technical Progress: A Cross-Industry Analysis (New York, 1982), 9–100Google Scholar; and Philip Seidenberg, “From Germanium to Silicon: A History of Change in the Technology of the Semiconductors,” in Goldstein and Aspray, eds., Facets: New Perspectives on the History of Semiconductors, 35–74.
10 Most of the authors listed above, with the exception of Seidenberg (who points to forces within firms and research labs as driving change in the industry), would say Department of Defense support has been a significant source of technological innovation in the industry, particularly in its infancy. The government bought 100 percent of all integrated circuits produced in 1962, for example. By 1965, the government's share had dropped to 72 percent, and by 1968, it was a mere 37 percent of the year's total shipment of integrated circuits. Levin, “The Semiconductor Industry,” 63. For more on the role of the military in technically advanced industries, see Mowery and Nelson, “Explaining Industrial Leadership,” in Mowery, and Nelson, , Sources of Industrial Leadership; Smith, Merritt Roe, ed., Military Enterprise and Technological Change: Perspectives on the American Experience (Boston, 1985Google Scholar).
11 Christophe Lecuyer, “Fairchild Semiconductor and Its Influence,” in Lee et al., The Silicon Valley Edge, 158.
13 I am reminded of an observation made by Harold Livesay in the pages of this journal in 1989. Business history's overriding attention to bureaucracy and organizational synthesis, wrote Livesay, “has in recent years, acted much like a neutron bomb, wiping out the people while leaving the buildings intact.” Livesay, Harold C., “Entrepreneurial Dominance in Businesses Large and Small, Past and Present,” Business History Review 63 (Spring 1989): 5.CrossRefGoogle Scholar
14 Individual entrepreneurs working in Silicon Valley at the end of the twentieth century have recently begun to receive some recognition from scholars. See, for example, Chong-Moon Lee, “Four Styles of Valley Entrepreneurship,” in Lee et al., The Silicon Valley Edge, 94–123.
15 In addition to other cited sources, the author's interviews with Jim Gibbons (2 June 1999), Gordon Moore (30 March 1999), Harry Sello (25 Jan. 1999), and two anonymous Fairchild founders informed this paragraph. Good discussions of Shockley Transistor Labs can be found in Lecuyer, “Fairchild Semiconductor,” in Bassett, “New Organizations,” and in Riordan, Michael and Hoddeson, Lillian, Crystal Fire: The Birth of the Information Age (New York, 1997Google Scholar).
16 Germanium was the preferred semiconductor substrate at the time Shockley began his company. Silicon's higher melting point made it harder to work with than germanium, but silicon (the basic ingredient in sand) is far more abundant, and devices made from it were more reliable than germanium devices. Riordan and Hoddeson, Crystal Fire, 207–8, 221. In the diffusion process, a semiconductor is cooked in a furnace containing appropriate impurities that then seep into the semiconductor in much the same way that hickory flavor seeps into meat cooked in a barbecue pit. Barbecue analogy from Reid, T. R., The Chip (New York, 1985), 73–4Google Scholar.
17 For more on “Hot Minds,” see Crystal Fire; “Secrets of the Mind,” Newsweek, 6 Dec. 1954, 72. Shockley's work on empirical measures for intelligence later caused him problems, when he used his results to argue that blacks were intellectually inferior to whites.
18 For example, he briefly barred Jean Hoerni, a brilliant experimentalist, from lab work and assigned him instead to do the “dog work” of calculating diffusion curves in an apartment Shockley had rented for him not far from the lab. Shockley also believed in public firings; he would gather the group and tell one unsuspecting man to clear out his desk—with no advance warning. Fairchild Founder A, interview by author, 27 Jan. 2000. (A handful of the more than two dozen people I interviewed requested anonymity.)
19 On the shifts from research to product development: Jim Gibbons, interview by author, 2 June 1999; Fairchild Founder B, interview by author, 19 March 1999; Fairchild Founder A, interview by author, 27 Jan. 2000. A diode functions much like a dam or a door, opening at times to let current pass and firmly impeding the flow of current at other times. (A circuit designer determines the conditions under which the diode allows current to pass.) Shortly after Shockley was awarded the Nobel Prize for Physics in November 1956, he began work to build a diode with four layers that would be diffused P–N-P–N. “Inventor Cites Use in Computer,” Electronic News, 24 Feb. 1958, 1. For more on the four-layer diode, see Riordan and Hoddeson, Crystal Fire, 267–8.
20 Gordon Moore recalls that he and Noyce wrote a letter to Shockley making these points. Gordon Moore, interview by author, 30 March 1999. I have not found the letter in Shockley's voluminous collection in the Stanford Special Collections. For more on these manufacturing difficulties with the four-layer diode, see Riordan and Hoddeson, Crystal Fire, 267. For more on the limited customer base for the product, see Bassett, “New Technology.”
21 Says Harry Sello, who left Shockley for Fairchild Semiconductor six months after Noyce did: “They could have done that [built the silicon transistor] as easy as falling off a log at Shockley Laboratories if the old man would only have taken the lid off.” Harry Sello, interview by author, 25 Jan. 1999. Lecuyer writes that the group had managed to build experimental mesa transistors at Shockley. Lecuyer, “Making Silicon Valley,” 154.
22 The story of the contact between the group of seven and Hayden, Stone has been recounted in almost every source listed in footnote 3, though at times it is mistold to include Noyce as one of the original dissidents. Good recountings are by Riordan and Hoddeson, Crystal Fire, and Lecuyer, “Making Silicon Valley” and “Fairchild Semiconductor.” The full text of the letter can be found in “The Founding Documents,” Forbes ASAP, 29 May 2000, unpaginated.
23 Arthur Rock, interview by author, 23 Feb. 1999.
24 Gordon Moore: “We got together for an evening and they say [sic], ‘Hey, you don't want to find a company to hire you. What you want to do is set up your own company.’ Our own company, yeah, OK. That way we won't even have to move. So that was the entrepreneurial spirit that drove the formation of Fairchild.” Gordon Moore, interview by Alan Chen, 9 July 1992. Intel Archives.
25 Fairchild Founder A, interview by Christophe Lecuyer on 6 July 1996. Personal communication from Lecuyer to author, 11 Nov. 1999.
26 Gordon Moore, interview by author, 30 March 1999.
27 Entry dated 5 June 1957, Empire Notebook, Shockley Papers, Stanford Special Collections (henceforth SSC).
28 Michael F. Wolff, “The Genesis of the Integrated Circuit: How a Pair of U. S. Innovators Brought Into Reality a Concept that was on the Minds of Many,” IEEE Spectrum, Aug. 1976, 49. Various versions of this story exist, including an official straw poll, but Wolff writes that Noyce confirmed this version of the story, albeit “with some embarrassment.”
29 Entry labeled 6 Jun; entry labeled “Impressions from AOB [Arnold O. Beckman], call Thurs PM 6 Jun,” Empire Notebook, Shockley Papers, SSC. Shockley's notes from the period of most intense turmoil in the company appear in a spiral-bound “Empire” brand notebook dated 23 May–10 June 1957, on the front of which he has scrawled, as if to remind himself: “Try to work it out for the benefit of everyone.” “Like you did at meeting, listening.”
30 Shockley recorded few conversations with any other lab employees at this time, with the exception of Smoot Horsley, who ran the four-layer diode operation and was not part of the disaffected group. Empire Notebook, Shockley Papers, SSC.
31 Gordon Moore, interview by author 30 March 1999.
32 Arthur Rock, interview by author, 23 Feb. 1999.
33 Fairchild Founder A, interview by Christophe Lecuyer on 6 July 1996.
34 Fairchild Founder A, interview by author, 27 Jan. 2000.
35 Statistics are for Oct. 1956–Oct. 1957. ”1958: Everybody's Doin' It,” and “The Transistor Emerges,” Electronic News 25th Anniversary Issue, 25 Jan. 1982 (sect. 2), 6, 23. See also, Alfred Cook and Bob Shephard, “Heavy Commercial Push Top Feature at Wescon,” Electronic News, 26 Aug. 1957, 1.
36 Arthur Rock, interview by author, 23 Feb. 1999.
37 Gordon Moore, interviewed by Alan Chen, 6 Jan. 1993, transcript. Intel archives.
38 “Multifarious Sherman Fairchild,” Fortune, May 1960, 170; “Sherman Fairchild—Man of Few Miscalculations,” Electronic News, 13 Sept. 1965, 8.
39 “Multifarious Sherman Fairchild,” Fortune, 170.
40 Sources for Fairchild statistics ca. 1957 are Fairchild Camera and Instrument Annual Report 1957; “Man in a Hurry,” Forbes, 15 Nov. 1959, 37; and “Fairchild Camera—Portrait of a Growth Company,” Financial World 118, 12 Sept. 1962, 10–11.
41 “Rough Road to Growth,” Forbes, 15 June 1965, 30.
42 In industries such as automobile manufacturing, oil refining, and tobacco production, many automated operations and testing functions either already used semiconductors or were investigating how to do so. Ken Stein, “Experience in Field is Opening Markets,” Electronic News, 17 Feb. 1958, 1.
43 Fairchild Camera and Instrument Annual Report 1957. In Noyce's words, Fairchild management was “primed and eager to go along if [the Shockley defectors'] ideas seemed practical.” “Fairchild Semiconductor Corporation: Company Profile,” Solid State Journal, Sept./Oct. 1960, 1.
44 Richard Hodgson, interview by author, 19 May 1999.
45 Fairchild Founder B, interview by author, 19 March 1999.
46 Contract between “the California Group” and “Fairchild Controls,” 19 Sept. 1957, Shockley Papers, SSC. How Shockley got a copy of the contract is a mystery.
47 Gordon Moore, interview by author, 30 March 1999.
48 Shockley Labs had been one of the few firms in the country working in silicon at the time the eight defectors decided to build silicon transistors. Even Bell Labs, a hotbed of silicon research in the late-1950s, nonetheless built almost all of its transistors from germanium. Philip Seidenberg, “From Germanium to Silicon,” 37.
49 Letter from L. N. Duryea to Eirckson, Wright, Hanafin, and Steinmeyer, 28 May 1959, Shockley Papers, SSC.
50 Of course, corporate genealogy is not destiny, and Semiconductor's descent from Shockley did not guarantee success. Witness, for example, the less-than-auspicious career of Knapic Electro-Physics, a company founded by Shockley employee Dean Knapic a few months after the Fairchild Semiconductor team defected. Daily Palo Alto Times, 18 Dec. 1957. See also, “Transamerica Plan Gets FRB Approval,” San Francisco Chronicle, 18 Dec. 1957.
51 Certificate of Winding Up and Dissolution, 18 April 1960, Shockley Papers, SSC.
52 There were seven voting trustees: Carter, Hodgson, and two other Camera and Instrument senior managers; Bud Coyle from Hayden, Stone; and Noyce and Eugene Kleiner.
53 Contract between “the California Group” and “Fairchild Controls,” 19 Sept. 1957, Shockley Papers, SSC. See also letter from Bob Noyce to employees, “Fairchild Semiconductor, 1957–1977” (booklet of reproduced items pertaining to the first twenty years of Fairchild Semiconductor's existence), SSC. Semiconductor's finances were kept entirely separate from those of Camera and Instrument, resulting in nice accounting benefits for the parent firm. Rather than spending the $1.4 million itself—and taking a $1.4 million hit against earnings—Camera and Instrument could account for the expenditures piecemeal, so that, for example, less than $400,000 of Semiconductor-related expenses were filed against earnings in 1958. This fancy footwork translated to an 80-cents-per-share savings to Camera and Instrument shareholders. Fairchild Camera and Instrument Annual Report 1958.
54 Richard Hodgson to Bob Noyce, 2 Oct. 1957, SSC; Bichard Hodgson, interview by author, 19 May 1999.
55 This creation of a new type of organization, by combining in a new way resources available to others, is, of course, the essence of Schumpeter's definition of entrepreneurship.
56 “8 Leave Shockley to Form Coast Semiconductor Firm,” Electronic News, 20 Oct. 1957.
57 Richard Hodgson, interview by author.
58 Jim Gibbons, Gordon Moore, Harry Sello, and Richard Hodgson commented that Noyce did nothing deliberate to attract attention to himself. Nonetheless, Sello, who worked with Noyce at both Shockley and Fairchild, told a revealing story of Noyce's guest appearance on a public-access television show Sello hosted, called “This Week in Science.” In the course of the appearance, Sello says, Noyce “stepped in front of me. And I'm the damn lead on the show. He cut my lines. He upstaged me all over the place. … I made a crack about, ‘Do you see why he's president of the organization and I'm not?’ He burst out laughing.” Harry Sello, interview by author, 25 Jan. 1999.
59 Paul Hwoschinsky, interview by author, 25 March 1999. The best discussion of charisma in economic organizations is Weber, Max, “The Sociology of Charismatic Authority,” in Gerth, H. H. and Mills, C. W., translators and eds., From Max Weber: Essays in Sociology (New York, 1946Google Scholar).
60 Fairchild Founder A, interview by author, 27 Jan. 2000.
61 See, for example, Gordon Moore's discussion of Noyce's suggestion to use nickel plating in an attempt to keep transistor junctions from leaking. Gordon Moore, interviewed by Alan Chen, 6 Jan. 1993, transcript, Intel archives.
62 Hoerni, quoted in Bob Ristelhueber, “Noyce Remembered: Unusual Ideas, Unusual Approaches,” Electronic News, 11 June 1990, 4. While at Fairchild, Hoerni developed the planar process, which made batch production of semiconductors possible. The planar process also was the foundation for Noyce's conception of the integrated circuit, which combined in a single device all the components needed to perform an electronic function.
63 One measure of the importance of the integrated circuit is the award of the 2000 Nobel Prize in Physics to Texas Instrument's Jack Kilby, who is generally acknowledged to be the “coinventor,” with Noyce, of the integrated circuit. It is safe to say that if the prize were awarded posthumously, Noyce would have shared it. “Jack S. Kilby and Robert Noyce are both considered as the inventors of the integrated circuit,” reads the Nobel announcement of Kilby's award, http://nobel.sdsc.edu/announcement/2000/phyinfoen.html (accessed 1 Dec. 2000). Noyce's integrated circuit patent is #2981877; Kilby's is #3138743; Interference #92,841; Case 416 F2d 1391 (1969).
64 In May 1959, the eight founders agreed that Noyce, originally head of Research and Development, should serve as general manager after Ed Baldwin, the firm's first general manager, announced that he and eight senior operations people (including five engineers) were decamping to form a rival semiconductor operation. R. Dale Painter, “Seek to Settle Suit on Rheem Semiconductor,” Electronic News, 14 March 1960, 12.
65 Gordon Moore, interview by author, 30 March 1999.
66 The math is roughly as follows. If only half of the transistors on a given wafer are good, then putting together any two transistors yields chances of only one in four that the combination works; in the case of four transistors, only one-sixteenth are good. There was talk that with twenty or thirty transistors in a given circuit, yields would be so low that each functional circuit would have to cost a fortune. (Even a best-case scenario of 90 percent of transistors actually working resulted in 12 percent overall yields of 20-transistor circuits.) For more on this point, see Robert Noyce, “Machine that Changed the World,” Intel Archives; Kilby, “Invention of the Integrated Circuit,” 652.
67 The market Noyce and Moore had anticipated took longer to develop than either had predicted. Ultimately, the company felt compelled to introduce a series of drastic price cuts that made it cheaper for a customer to buy an integrated circuit than to buy the individual components and connect them himself. This pricing scheme should be understood as an important element in the “invention” of the integrated circuit: in the language of the history of technology, cost was an essential factor in the “social construction” of the integrated circuit. For a good introduction to social construetivist theories that technologies are not autonomous technical developments, but rather a product of a variety of political, economic, and social influences, see Hughes, Thomas, Bijker, Wiebe E., and Pinch, Trevor J., The Social Construction of Technological Systems (Cambridge, Mass., 1987Google Scholar); Hounshell, David, From the American System to Mass Production, 1800–1932: The Development of Manufacturing Technology in the United States (Baltimore, 1984Google Scholar); and MacKenzie, Donald A., Inventing Accuracy: A Historical Sociology of Nuclear Missile Guidance (Cambridge, Mass., 1990Google Scholar). An interesting (and somewhat parallel) discussion of “technological trajectories” and the ways that a certain technological approach becomes the “dominant design” is by Nelson, Richard R., “The Co-Evolution of Technology, Industrial Structure, and Supporting Institutions,” in Dosi, Giovanni, Teece, David J., and Chytry, Josef, eds., Technology, Organization, and Competitiveness: Perspectives on Industrial and Corfiorate Change (New York, 1998), 319–35CrossRefGoogle Scholar. For an interesting case within the semiconductor industry, see William Aspray, “The Social Construction of the Microprocessor; A Japanese and American Story,” in Facets: New Perspectives on the History of Semiconductors, 215–67.
68 Wolff, “The Genesis of the Integrated Circuit,” 53.
69 Arthur Rock to partners of Hayden, Stone, and Co., reprinted in “The Founding Documents.”
70 “Multifarious Sherman Fairchild,” 171. The Western Union telegram notifying Fairchild Semiconductor of the buyout is reproduced in “Fairchild Semiconductor, 1957–1977,” SSC.
71 Fairchild Camera and Instrument Annual Report 1959.
72 Fairchild Camera and Instrument Annual Report 1960, 1961. “Strong Position of Firm Cited by Fairchild Semiconductor VP,” Electronic News, 20 March 1961, 16.
73 Fairchild Camera and Instrument Annual Report 1961.
74 Fairchild Camera and Instrument Annual Report 1962.
75 The stock rose faster than any other on the New York Stock Exchange. “What Made a High Flier Take Off at Top Speed,” Business Week, Oct. 1965, 118–20; “Exchange Calls FC&I Pacer,” Electronic News, 7 Feb. 1966.
76 The patents in question covered a method of building integrated circuits using the planar process. For more on the importance of licensing and cross-licensing in the semiconductor industry, see Langlois and Steinmueller, “The Evolution of Competitive Advantage in the Worldwide Semiconductor Industry,” 19–78.
77 Braun and Macdonald, Revolution in Miniature, 78.
78 The difficulties of transferring technologies within organizations are unfortunately familiar to many industries. See, for example, Graham, Margaret B. W., The Business of Research: RCA and the VideoDisc (Cambridge, U.K., 1986Google Scholar). Nonetheless, the “black magic” problems (detailed below) faced by the semiconductor industry were such highly specialized variations on this common theme that I consider them an issue specific to the semiconductor industry.
79 We now know that many of these problems had to do with contamination. The Technical and Progress Reports in the SSC are full of examples of the problems entailed in moving from development to manufacturing. To take just one: “The 1210 is behaving in the normal fashion for any device with drastically increased production. The first groups of runs were lost because of a too high base predep [predeposition] temperature. This has been tracked down and corrected, but because of the time lag between base predep and emitter diffusion, close to two weeks work was lost. A problem that has come up since then is the presence of a large percentage of soft units in one run. This has been very uncommon in the 1210.…” The writer later says that several problems stem from the “transition to Mt. View,” where Semiconductor's manufacturing facilities were housed. “Progress Report—Transistor Development Section, 1 Jan. 1961,” Box 5, File 9, Fairchild Research and Development Division, Technical Reports and Progress Reports, M1055, SSC.
80 “Letter from Moore and Grinich to Noyce, 8 Feb. 1961,” Box 5, File 11, Fairchild Research and Development Division, Technical Reports and Progress Reports, M1055, SSC.
81 “Trimmed oxide now helps, even though a year ago it did not.” “R&D Progress Report from Moore and Grinich to Noyce, 11 Aug. 1961,” Box 6, File 7, Fairchild Research and Development Division, Technical Reports and Progress Reports, M1055, SSC.
82 “Letter from Moore and Grinich to Noyce, 8 Feb. 1961,” Box 5, File 11, Fairchild Research and Development Division, Technical Reports and Progress Reports, M1055, SSC.
83 “R&D Progress Report from Moore and Grinich to Noyce, 11 April 1961,” Box 6, File 3, Fairchild Research and Development Division, Technical Reports and Progress Reports, M1055, SSC.
84 See, for example, “Letter from Eugene Kleiner to Gordon Moore, 1 Dec. 1960,” Box 5, File 5, Fairchild Research and Development Division, Technical Reports and Progress Reports, M1055, SSC. In this letter, Kleiner explained that staff in Palo Alto were accusing people from Mountain View of taking Palo Alto's supplies of paper and pencils.
85 “R&D Progress Report from Moore and Grinich to Noyce, 11 April 1961,” Box 6, File 3, Fairchild Research and Development Division, Technical Reports and Progress Reports, M1055, SSC.
86 The spin-off phenomenon was particularly pronounced between 1961 and 1963, and again in the period from 1968 to 1969. Similar forces drove both generations of startup companies. In particular, each generation was founded around the promise of a specific technology—in the early 1960s, the planar process and integrated circuit; in the second wave, large-scale integration (LSI) techniques.
87 Kraus, Jerome, “An Economic Study of the U.S. Semiconductor Industry” (unpublished New School for Social Research diss., 1973), 110Google Scholar. In 1958 and 1959, electronics stocks rose roughly three times as fast as the Dow Jones industrial average: 50 percent vs. 38 percent in 1959; 60 percent vs. 30 percent in 1958. Stuart Gellman, “Industry Still Flying High in the Market,” Electronic News, 28 Dec. 1959, 1. See also, “Transitron Offer Sparks Broker Deluge,” Electronic News 25th Anniversary Edition, 25 Jan. 1982 (sect. 2), 28. Alfred D. Cook, “Letter from the Editor,” Electronic News, 15 Feb. 1960, 2.
88 Kraus, “An Economic Study of the U.S. Semiconductor Industry.”
89 Interesting contemporary commentary on the phenomenon can be found in “Ex-Emplove[e]s File $700,000 Counter Claim,” Electronic News, 30 Jan. 1961; “Silicon Transistor Sues Ex-Emplove[e]s, Seeks $1 Million,” Electronic News, 23 Jan. 1961; “Diotran Pacific Formed by Four In Palo Alto, Cal,” Electronic News, 6 March 1961; “Melapar Sues Scope, Others for $500,000,” Electronic News, 27 July 1961, 1. See also reports on Computer Dynamics Corp., a Maryland firm which grew from 5 to 100 in the space of a year: “5 Man Firm Grows to 100 in First Year,” Electronic News, 7 Jan. 1963, 49. Estimate of the number of new firms is from “The Semiconductor Industry: Mayhem and Millionaires,” Electronic News, 25 Jan. 1982 (sect. 2), 16.
90 Fairchild Founder B, interview by author, 19 March 1999; Charlie Sporek, interview by author, 16 Feb. 1999; Fairchild Camera and Instrument Annual Report 1959. At year's end, 52,250 shares that had been authorized for issue under the stock option plan had not been issued. Fully 26,000 of these shares had been authorized during the year 1959.
91 Richard Hodgson, interview by author, 19 May 1999. Stock options had gained popularity in the 1950s, when firms began granting them to members of the management team. By the end of that decade, analysts were arguing that technical and scientific men, not just managers, should receive options, and by the early 1960s, leading scientists in the electronics industry expected some form of stock option as part of their compensation package. “You can't get anyone today,” one electronics-firm manager complained to Business Week, “unless you can offer him a stock option.” “Business Week Reports on Semiconductors,” Business Week, 26 March 1960, 110; R. W. Johnson, “Stock Options—Their Place in Engineering Management,” Western Electronic News, May 1957, 8.
92 Charlie Sporek, interview by author, 16 Feb. 1999.
93 Gordon Moore has called Noyce “a man of many ideas, some of them good” and has said that Noyce never solved problems in a straightforward, systematic way. Gordon Moore, interview by author, 30 March 1999.
94 Harry Sello, interview by author, 25 Jan. 1999.
95 Richard Steinheimer (graphic artist), interview by Rob Walker, SSC, 25 May 1999.
96 Richard Levin, “The Semiconductor Industry,” 54. By Levin's estimate, Semiconductor labs produced 21 percent of the device structure innovations, 18 percent of new product innovations, and 12 percent of new process innovations. Levin estimates that 10 percent of the major innovations during the discrete device era, 1950–1960, can be traced to Fairchild Semiconductor—an impressive showing, given that Fairchild did not enter the business until the end of 1957.
97 Frank Wanlass quoted in George Rostky, “The 30th Anniversary of the Integrated Circuit: Thirty Who Made a Difference,” 1988. SSC; Bassett, “New Technology,” 99–100.
98 See Lecuyer, “Making Silicon Valley,” 176–7. A representative advertisement appears in “A Briefing on Integrated Circuits,” television show broadcast in October 1967, courtesy of Harry Sello.
99 Walter Matthews, “Shift Semicon Sales Setup at Fairchild,” Electronic News, 6 April 1964, 1.
100 “A Briefing on Integrated Circuits,” television show broadcast in October 1967, courtesy Harry Sello. “Fairchild ‘Special’ Aimed at Select Group,” Broadcasting, 2 Oct. 1967, 35. The article calls the special “a landmark event in television the first of its kind in TV history.” The Fairchild annual report for 1967 notes that the show was carried by 32 stations and “viewed by an estimated 2 million persons. This is [the] first known use of commercial TV to teach a technical subject.”
101 “Fairchild Starts Hong Kong Venture,” Electronic News, 8 April 1963, 6; Walter Mathews, “Hong Kong Semicon Plant for Fairchild,” Electronic News, 17 June 1963, 34. Charlie Sporck, interview by author, 16 Feb. 1999.
102 Charlie Sporck, interview by author, 16 Feb. 1999; Lecuyer, “Making Silicon Valley,” 244. Many of the sources mentioned in footnote 9 address the issue of automation in the semiconductor industry.
103 Other firms included Continental Device Corp., General Electric, and ITT Semiconductor. Don Hoefler, “Hit 'Em Where They Ain't,” Electronic News, 15 Jan. 1968, 1.
104 “Man Behind the News,” Electronic News, 3 Dec. 1962.
105 Noyce, interviewed by Herbert Kleiman, SSC. Some scholars have argued that military work was essential to the development of a culture of continuous innovation in the electronics industry. The military often was more interested in funding the development of state-of-the-art prototype devices (each of which required a great deal of research and development work) than in buying mass quantities of less advanced technology. The theory posits that the existence of a deep-pocketed customer interested in funding innovation after innovation generated a business culture focused on R&D rather than manufacturing. The irony here, of course, is that the innovation-over-manufacturing culture could well describe Fairchild, which nonetheless had very little to do with direct government work. For an introduction to these arguments, see Markusen, Ann, Hall, Peter, Campbell, Scott, and Deitrick, Sabina, The Rise of the Gunbelt: The Military Remapping of Industrial America (New York, 1991Google Scholar).
106 Noyce, interview by Kleiman. In their interviews with the author, Charlie Sporck and Gordon Moore confirmed the distaste for direct government contracts at Semiconductor.
107 In early 1963, for instance, Moore estimated that while government contracts represented “only about 5% of our total laboratory effort [in 1962], they have given us invaluable experience.” “Research and Development Progress Report of 1962, 24 Jan. 1963,” Box 7, File 6, Fairchild Research and Development Division, Technical Reports and Progress Reports, M1055, SSC.
108 Harry Sello recalls that at Semiconductor, “We carefully chose [government projects] so that the government needed the product and it was work that we needed also in order to promote the commercial aspects of the product.” Harry Sello, interview by author, 25 Jan. 1999.
109 Fairchild Camera and Instrument Annual Report 1960.
110 Lecuyer, “Making Silicon Valley,” chs. 3 and 4.
111 Charlie Sporck, interview by author, 16 Feb. 1999. See also Lecuyer, “Making Silicon Valley,” 177–95. Lecuyer points out that a contract from Autonetics required the company to “reinforce [its] manufacturing discipline … document [its] manufacturing processes … [and] build high-reliability lines using [specific production] techniques” (189).
112 See Charlie Sporck, interview by author, 16 Feb. 1999.
113 Lecuyer, “Making Silicon Valley,” 239. See 237–47 for a good general discussion.
114 “You'd think these two men [Bay and Sporck] were going to fight each other on the floor,” Richard Hodgson recalls, adding that Noyce would usually just sit back and let the arguments run their course. Richard Hodgson, interview by author, 19 May 1999.
115 To take but one example of volume production: in 1965, integrated circuits of a sort called complementary-transistor logic (CTL) were shipped in quantities of a half-million units or more. Fairchild Camera and Instrument Annual Report 1965.
116 Charlie Sporck, interview by author, 28 Dec. 2000.
117 For quite a while, Semiconductor had built its own test equipment. The instrumentation division was formed to commercialize this equipment, and it also incorporated resources from Dumont Laboratories, an early Fairchild acquisition that had been plagued with troubles for years. Fairchild Camera and Instrument Annual Report 1965; Don Hoefler, “Shakeup in Sunnyvale,” Electronic News, 16 Feb. 1970.
118 Fairchild Camera and Instrument Annual Report 1965, 1966. In 1965, Fairchild Camera and Instrument had 11,545 employees; in 1966, the employee base was 13,591.
119 Fairchild Camera and Instrument Annual Report 1966. The company had roughly 2.6 million shares outstanding.
120 James Lydon, “Recruiting Fever Mounts in Microelectronics Field,” Electronic News, 25 July 1966, 1.
121 For an excellent discussion of tensions between Semiconductor's Palo Alto and Mountain View facilities in the late 1960s, particularly in connection with MOS developments, see Bassett, “New Technology,” 256–70.
122 “R&D Progress Report from Moore and Grinich to Noyce, 11 June 1962,” SSC.
123 Bassett, “New Technology,” 208. Bassett's dissertation focuses on the development of MOS technology at Fairchild Semiconductor and elsewhere.
124 For example, in February 1963, Semiconductor announced it would soon bring to market an MOS transistor, only three weeks later to issue a statement that “Fairchild Semiconductor's field-effect metal-oxide semiconductor transistors are still in the laboratory and are not for sale. [The company] does not know when they will be released from the laboratory.…” Walt Bohne, “New Fairchild Transistor Seen at 1/3 Usual Cost,” Electronic News, 3 Feb. 1964, 32; “Fairchild Field Effect Transistors Still in Lab,” Electronic News, 26 Feb. 1964, 37.
125 “Dr. Noyce candidly admitted that Fairchild's history with the metal-oxide-silicon technology has been rather erratic. While the company got into it early under the impetus of Frank Wanlass, when he departed for General Micro-Electronics, and later general Instrument Corp., Fairchild's MOS effort became spotty.” Don Hoefler, “FC&I, Mountain View, Breathes Easier,” Electronic News, 30 Oct. 1967. For an excellent discussion of MOS at Fairchild Semiconductor, see Bassett, “New Technology,” chs. 4 and 5.
126 Charlie Sporck, interview by author, 28 Dec. 2000.
127 Don Hoefler, “FC&I Profit Dip on ICs,” Electronic News, 21 Nov. 1966, 12.
128 Hoefler, “FC&I Profit Dip on ICs”; Walter Matthews, “Geographic Expansion Set by Fairchild,” Electronic News, 19 July 1965.
129 Roger Borovoy, interview by author, 27 Jan. 1999.
130 Hoefler, “FC&I Profit Dip on ICs,” 12. Camera and Instrument never broke down earnings by division, so it is impossible to determine the exact impact of the troubles at Semiconductor on the division's bottom line.
131 Fairchild Camera and Instrument Annual Report 1966.
132 Robert Noyce to Sherman Fairchild, 25 June 1968. Camera and Instrument did not break down earnings by division, but Time estimated that Semiconductor was responsible for 98 percent of the parent company's profits. “Mighty Miniatures,” Time, 4 March 1965, 93–4.
133 Fairchild Camera and Instrument Annual Report 1966.
134 “What Made a High Flier Take Off at Top Speed,” 118–22.
135 “Nat'l Semiconductor Moving, Realigning Top Management,” Electronic News, 6 March 1967, 54. Among those who left with Sporck, or shortly thereafter to join him, were Floyd Kvamme, marketing manager for ICs; Pierre Lamond, IC production manager; Roger Smuller, manufacturing manager for ICs, Fred Bialek, overseas operations manager for microcircuits; and Don Valentine, director of marketing.
136 Hoefler, “FC&I Profit Dip on ICs.”
137 Charlie Sporck, interview by author, 16 Feb. 1999.
139 “I suppose I essentially cried when he left,” Noyce told a reporter. “Working with people that you're fond of, then having them break apart, was I would almost say devastating.” Malone, Big Score, 108.
140 Gordon Moore, in an interview with the author, recalls that Noyce first offered the job to him. Bay laid off 150 people at the Mountain View manufacturing facility and granted marketing a great deal of power over key decisions, such as what products to make, and when. For example, Jerry Sanders, the new marketing manager, spearheaded the introduction of a new product every week for all of 1968. “The Fight That Fairchild Won,” 106; “Fairchild Semicon Lays off 150 at Coast Plant,” Electronic News, 22 Jan. 1968, 45; “The Semiconductor Industry: Mayhem and Millionaires,” 16.
141 Robert Noyce, “WGBH—‘The Machine That Changed the World,’” Tape F7, 6–7, Interview Transcript, Intel Archives. In an interview with the author, Sporck recalled that Noyce never even bothered trying to talk him out of leaving. Charlie Sporck, interview by author, 16 Feb. 1999.
142 “Turning a Science into an Industry,” IEEE Spectrum, Jan. 1966, 101.
143 This figure takes into account a three-for-two stock split during 1967. The additional shares followed on the granting of 215,525 new options—an enormous increase (even taking into account the stock split) over the previous year's 73,400, and the 1965 grant of 5,300. Fairchild Camera and Instrument Annual Report 1965, 1966, 1967.
144 Richard Hodgson, interview by author, 19 May 1999.
145 “The Fight That Fairchild Won,” 106. See, for example, James Lydon, “Five Leave FC&I Div. For ITT Portugal,” Electronic News, 27 March 1967; “2 Leave FC&I, Form Firm,” Electronic News, 15 May 1967, 39.
146 “The Fight That Fairchild Won,” 100.
147 “Coast Firm Unplugs Jam in Output,” Electronic News, 20 March 1967, 12.
148 See, for example, “Paying the Piper,” Forbes, 15 Nov. 1967, 65; “FC&I Head Resigns; Earnings Plummet,” Electronic News, 27 Oct. 1967, 2.
149 Don Hoefler, “FC&I, Mountain View, Breathes Easier,” Electronic News, 30 Oct. 1967.
150 “Paying the Piper,” 65; “Carter Resigns; Earnings Plummet,” Electronic News, 23 Oct. 1967.
151 Alfred D. Cook, “Sherman Fairchild's July 4th Fizzles,” Electronic News, 8 July 1968, 1.
152 “Operations [at the semiconductor division] were profitable in 1967, although increasing costs, expansion of operations with resulting production difficulties in certain new devices, and lower prices resulted in narrower margins,” Fairchild Camera and Instrument Annual Report 1967.
153 Cook, “Sherman Fairchild's July 4th Fizzles,” 1.
154 Carter attempted to rally a small group of directors to defend his acquisitions strategy when the rest of the board wanted to divest themselves of the losing operations. When Carter's rally failed, he quit before he could be fired. Journalist Don Hoefler claimed that Noyce forced Carter's departure. Hoefler recalls that a few days after Carter left, Noyce turned to the reporter and said (with an “intense countenance”), “When you set out to kill the king, you'd better kill him dead” Don C. Hoefler, “Captains Outrageous,” California Today, 28 June 1981, 42.
155 Charlie Sporck, interview by author; Gordon Moore, interview by author; Roger Borovoy, interview by author.
156 Robert Noyce to Sherman Fairchild, 25 June 1968. In author's possession.
157 Gordon Moore, interview by Rob Walker, SSC.
158 “Semiconductor was the company,” Noyce said, “but they [Camera and Instrument] insisted on treating it as just another division.” Don Hoefler, “Dr. Noyce Happy Doing His Thing,” Electronic News, 28 Oct. 1968, 1.
159 “The Fight That Fairchild Won,” 112. This version of events is supported by details in “Musical Chairs,” Electronics, 19 July 1968, 45.
160 Lester Hogan, interview by Rob Walker, SSC, 22 Aug. 1995.
162 “Where the Action is in Electronics,” Business Week, 4 Oct. 1969, 90; Malone, Big Score, 124.
163 The one exception was marketing VP Tom Connors. “War Drums Start to Beat in Phoenix,” Electronic News, 19 Aug. 1968, 12. In two years, some sixty former Motorola employees caught what one wag called “the Motorola-to-Fairchild Express” to California. Stanford Erickson, “How Hogan Rescued Fairchild,” International Management, July 1970, 22; Don Hoefler, “Underground Express to California,” Electronic News, 9 Sept. 1968, 2.
164 “Fairchild Camera Formalizes Base Location to California,” Electronic News, 30 Sept. 1968, 70.
165 Lester Hogan, interview by Rob Walker, SSC, 22 Aug. 1995.
166 Although Noyce was nominally president and Moore vice president, both men insisted that the power division was equal. Nilo Lindgren, “Building a Rational Two-Headed Monster: The Management Style of Robert Noyce and Gordon Moore,” Innovation, n.d., but clearly 1970. Courtesy Regis McKenna.
167 Intel Annual Report 1979.
168 http://www.intel.com, accessed 28 Jan. 2001.
169 In the twenty years following the bombing of Pearl Harbor, over 400,000 new jobs were created in Santa Clara County, many at the Peninsula's booming electronics companies. Between 1960 and 1975, while the nation's employment rate grew 46 percent and California's 65 percent, Santa Clara County's increase was 156 percent. Annalee Saxenian, “The Genesis of Silicon Valley,” in Hall, Peter and Markusen, Ann, Silicon Landscapes (Boston, 1985), 20–34Google Scholar. By 1965, firms specializing in semiconductor manufacture had supplanted vertically integrated electronics firms as the leading U.S. merchant semiconductor manufacturers. Langlois and Steinmueller, “The Evolution of Competitive Advantage in the Worldwide Semiconductor Industry,” 33.
170 Chandler, Alfred D. Jr., “The Information Age in Historical Perspective,” introduction to Chandler, Alfred D. Jr. and Cortada, James W., eds., A Nation Transformed by Information: How Information Has Shaped the United States from Colonial Times to the Present (New York, 2000), 31Google Scholar.
171 By the end of Noyce's tenure at Fairchild, one reporter could write that the company “seemed dedicated to technology for its own sake. … At Fairchild, it almost seemed that no engineer wanted to be in production.” “Musical Chairs,” 45. Erickson, “How Hogan Rescued Fairchild,” 22. If, in the words of Raychem CEO Paul Cook, “what separates the winners and losers in innovation is who masters the drudgery,” at Semiconductor under Noyce, not enough people took responsibility for the drudgery. William Taylor, “The Business of Innovation: An Interview with Paul Cook,” Harvard Business Review (March–April 1990): 97–106.
172 Robert Noyce, interview with Mary Burt Baldwin, Intel Archives. Noyce's situation was far from unique. In a recent Harvard Business Review article, Robert Goffee and Gareth Jones point out that it is unusual for “great leaders” to be “great coaches.” More typical, they write, are leaders “whose distinctive strengths lie in their ability to excite others through their vision [rather] than through their coaching talent.” Robert Goffee and Gareth Jones, “Why Should Anyone Be Led By You?” Harvard Business Review (Sept.–Oct. 2000): 63. Howard H. Stevenson and David E. Gumpert make a similar distinction between “promoters” (entrepreneurs) and “trustees” (administrators). Stevenson and Gumpert, “The Heart of Entrepreneurship,” Harvard Business Review (March–April 2000): 84–94. A good case study is Cheape, Charles, Family Firm to Modern Multinational: Norton Company, a New England Enterprise (Cambridge, Mass., 1985Google Scholar), particularly his discussion of the arrival of the company's first professional manager, Robert Cushman.
173 Robert Noyce to Sherman Fairchild, 25 June 1968.