How Philips Built the Chip Industry

Two of the most strategically important companies in the world — ASML, the Dutch monopoly on the lithography machines that pattern every advanced chip, and TSMC, the Taiwanese foundry that manufactures the majority of them — have a single common ancestor: the Philips electronics company of Eindhoven. The story of how a Dutch light-bulb maker became, almost by accident, the seed of the modern semiconductor industry is one of the great untold stories of twentieth-century technology.

The light-bulb years: 1891 to 1960

Philips was founded in 1891 in Eindhoven, a small city in the southern Netherlands, by Gerard Philips and his father Frederik Philips. The company began as a manufacturer of carbon-filament incandescent lamps, in a region of the Netherlands that had previously been dominated by the textile industry. The choice of Eindhoven was partly practical (Gerard Philips had been a lighting engineer in the city) and partly strategic (the cheap labour and the central location in the Dutch-Belgian industrial belt). The first year was difficult, and the company nearly went bankrupt. The recovery came in 1894, when Gerard’s brother Anton Philips joined the business, and the two of them turned the company into Europe’s largest light-bulb maker by the early years of the twentieth century.[1, 12, 13]

The first decades of the twentieth century were the period of Philips’s industrial consolidation. The company expanded into radio valves in the 1920s, into radio receivers in the 1930s, and into television in the postwar period. In 1914, the company established the Philips Natuurkundig Laboratorium (NatLab) in Eindhoven, a research laboratory that was, for most of the twentieth century, the company’s secret weapon. The NatLab employed more than 2,000 researchers at its peak and produced a remarkable series of innovations, including the radio valve, the pentode, the audio cassette (with the Compact Cassette format standardised in 1962), the compact disc (developed with Sony in the late 1970s and released in 1982), the laser disc, and the Blu-ray format. The NatLab was, in effect, the Dutch Bell Labs: a self-contained industrial research campus that took ideas from basic physics through to commercial products, and that was the source of most of Philips’s competitive advantage for the better part of a century.[1, 4, 14]

The semiconductor work at Philips began in the early 1950s, with research at the NatLab on germanium and silicon transistors. The company founded Signetics in 1961 in Sunnyvale, California, as a joint venture with Fairchild Camera, to commercialise the 7400-series TTL logic family. Signetics was the first company to produce the 7400 series in volume, and the 7400 standard became — and remains — the lingua franca of digital logic. The Philips semiconductor group, headquartered in Nijmegen, became one of the largest in Europe, with a product line that included the chips for Philips’s own consumer products (radios, TVs, cassette players, CDs) and a merchant business that sold to third parties. The semiconductor group was, by the early 1980s, one of the most valuable assets in the Philips portfolio, but it was also the source of a strategic problem: it was too capital-intensive for a company whose core business was consumer electronics.[1, 5, 15]

The lithography programme: a money-loser is spun out

Philips’s involvement in lithography began in the 1960s, when the NatLab started to develop the optical step-and-repeat cameras that were the precursor of the modern stepper. The work was led by a small team of engineers, most of them physicists by training, and the early products were sold in small numbers to the Philips semiconductor group for the production of integrated circuits. The problem was that the lithography market was, by the early 1980s, dominated by the American companies (GCA, Perkin-Elmer) and the Japanese companies (Nikon, Canon), and the Philips stepper business was deeply unprofitable. The lithography group had spent more than a decade and tens of millions of dollars on the programme, and there was no realistic prospect of a return.[2, 16, 17]

The decision to spin out the lithography group came in 1983, when Philips, under the leadership of Wisse Dekker, decided to refocus the company on its core businesses (consumer electronics, lighting, healthcare) and divest peripheral activities. The lithography group was offered to ASM International, a small Dutch semiconductor equipment company based in Bilthoven, in a joint venture in which Philips would provide the technology and the engineers and ASM would provide the manufacturing and the commercial relationships. The new company, ASM Lithography, was established in 1984 with about 50 employees, a single product (the PAS 2000 stepper, an i-line machine), and a budget that was, even by the standards of the time, modest. The company’s first years were difficult, and the company nearly went bankrupt in 1988. The recovery came in 1991, with the launch of the PAS 5000 stepper, and the rest, as they say, is history.[2, 3, 17]

The decision to spin out the lithography group was, at the time, the right one. Philips was under pressure from shareholders to refocus on its profitable businesses, and the lithography group was an expensive R&D programme that, by any reasonable financial metric, was a money-loser. The spin-out was structured to give Philips a minority stake and a seat on the board, but the operating control passed to ASM. The genius of the decision, in retrospect, was not that Philips recognised the future of lithography (it did not), but that it gave the new company the freedom to take risks that the parent would never have allowed. The PAS 2000 stepper was, in the early years, an also-ran; the breakthrough came with the PAS 5000 and the immersion stepper, technologies that were developed by ASML without the constraints of the Philips corporate culture. The fact that ASML was free to bet the company on technologies that Philips had rejected was, in the end, the reason the bet paid off.[2, 3, 16]

The Taiwan connection: Philips, ITRI, and the birth of the foundry

The story of TSMC, like the story of ASML, runs through Eindhoven. In the late 1970s, the Taiwanese government, alarmed by the success of the Japanese and Korean semiconductor industries, decided to build a domestic chip industry. The instrument was the Industrial Technology Research Institute (ITRI), a government-funded R&D centre established in 1973. ITRI, under the leadership of Morris Chang (who joined as head of the Electronics Division in 1985), set up a pilot semiconductor project with a small team of engineers, most of them recently returned from graduate programmes in the United States. The project needed a partner with CMOS process technology, and the natural choice was Philips, which was one of the few European companies with a competitive CMOS process at the time.[6, 7, 18]

The Philips-ITRI technology transfer agreement was signed in 1983. Under the agreement, Philips provided the CMOS process technology (a 3-micrometre process, then state-of-the-art) and the engineering support to set up a pilot production line at ITRI’s Hsinchu campus. The Taiwanese engineers learned the process on the Philips equipment, and the pilot line began producing chips for the Taiwanese market in 1984. The project was, by any measure, a success: it created a generation of Taiwanese engineers who understood CMOS manufacturing, and it provided the basis for the next step, which was the spin-out of a commercial foundry. The first spin-out, United Microelectronics Corporation (UMC), was established in 1980 (technically before the Philips agreement, as a spinoff from ITRI’s earlier work), and the second, Taiwan Semiconductor Manufacturing Company (TSMC), was established in 1987 with Morris Chang as its first chairman and CEO.[6, 7, 19]

The Philips connection to TSMC was, however, deeper than the 1983 agreement. The first TSMC fabs were built using the Philips CMOS process, and the first TSMC engineers were trained at the Philips-ITRI pilot line. The Dutch-Taiwanese collaboration continued for years: Philips and TSMC operated a joint venture for several years in the 1990s, and the Dutch company retained a small equity stake in TSMC until 2007. The result was a remarkable transfer of know-how: the CMOS process that Philips had developed in the 1970s became, by the early 2000s, the dominant semiconductor manufacturing process in the world, and the Taiwanese engineers who had learned it at ITRI became the core of the most important chipmaker in history. Morris Chang himself often credited the Philips-ITRI partnership as the most important factor in the creation of TSMC, and the Dutch engineering culture that Philips brought to Taiwan — a willingness to invest in long-horizon technology, a preference for collaboration between the factory and the customer, and a respect for the manufacturing process itself — has been visible in TSMC’s operations ever since.[6, 7, 8]

The decline of Philips and the legacy of its children

Philips in the twenty-first century is a shadow of the company it was in the twentieth. The semiconductor group was sold in 2006 to a private-equity consortium led by KKR, becoming NXP Semiconductors. The consumer electronics business (TVs, audio, video) was sold off piecemeal in the 2000s and 2010s, with the audio and video business sold to Funai and the TV business spun off as a joint venture with TPV Technology. The lighting business was spun off in 2016 as Signify. The remaining company, headquartered in Amsterdam, is now a focused healthcare and personal-health business, with a small consumer-products division. The company that, in 1984, had the option to keep ASML and the option to keep its semiconductor group, and chose to divest both, is now worth, in 2024, less than a fifth of ASML alone.[9, 10, 11]

The decline of Philips is a story of strategic drift, of a company that was too successful in its core businesses (lighting, consumer electronics) to take seriously the businesses that were not yet core (semiconductors, lithography). The decision to spin out ASML was, in retrospect, the most consequential decision in the company’s history, and it was made because the lithography group was, in the early 1980s, a money-loser. The same logic applied to the decision to license the CMOS process to ITRI: the technology was not particularly valuable to Philips, but it was very valuable to a Taiwanese partner. Both decisions were, in the language of financial economics, value-creating for the recipients and value-neutral for Philips. Both were, in the language of strategic history, among the most consequential technology decisions of the late twentieth century. The lesson is not that Philips’s management was wrong; it is that the management of a successful incumbent is structurally unable to value a peripheral asset as much as a focused new entrant can.[3, 9, 10]

The legacy of Philips in the chip industry is, by any measure, extraordinary. The company did not invent the transistor (that was Bell Labs) or the integrated circuit (that was Kilby and Noyce), but it was, in the second half of the twentieth century, one of the most important semiconductor companies in the world. Its NatLab gave the industry the 7400 logic family, the compact disc, and the laser disc. Its spin-out, ASML, became the only company in the world that can build the most advanced lithography machines. Its technology transfer, via ITRI, became the foundation of TSMC, the most important chipmaker in history. The two most strategically important companies in the semiconductor industry — ASML in Veldhoven and TSMC in Hsinchu — are, in a real sense, the children of Philips in Eindhoven. The parent has not benefited from the success of its children. The children have inherited, in different ways, a piece of the parent’s culture. The story of how this happened, and of how the parent did not understand what it was giving away, is one of the most important and most instructive in the history of technology.[1, 2, 6, 7]

What is in it for the future

The Philips story has lessons that go beyond the history of one Dutch company. The first lesson is that the most strategically important technologies of an era are often the most underappreciated within the incumbents. ASML in 1984 was an unprofitable R&D group in a company whose profitable businesses were lighting and consumer electronics; the lithography programme was, by any reasonable financial metric, a poor use of Philips’s capital. The decision to spin it out was rational. The lesson is not that Philips should have kept the lithography group; the lesson is that the management of a successful incumbent is structurally unable to value a peripheral asset as much as a focused new entrant can, and that the policy implications — antitrust, industrial policy, technology transfer — are large.[3, 9, 10]

The second lesson is that strategic partnerships, even small ones, can have outsized long-term consequences. The 1983 Philips-ITRI agreement was a routine technology transfer, of a kind that happens thousands of times a year. The agreement was, at the time, a footnote in the Philips annual report. Forty years later, the agreement was the foundation of the most important chipmaker in the world, and the engineer who led the project (Morris Chang) was, by general consensus, the most important figure in the semiconductor industry of the late twentieth century. The lesson is that small, peripheral decisions can have outsized consequences, and that the policy framework should be designed to maximise the probability that such decisions are made.[6, 7, 8]

The third lesson is that engineering culture, not capital or technology, is the durable competitive advantage. ASML and TSMC are both, in 2024, among the most valuable companies in the world, and they are both, in important ways, Philips companies — not because they use Philips’s technology (they do not, in any direct way) but because they share a piece of Philips’s engineering culture. The Dutch willingness to invest in long-horizon technology, the Taiwanese respect for the manufacturing process, the preference for collaboration between the factory and the customer — these are the cultural elements that the children inherited from the parent. The lesson is that the things that cannot be bought or sold — the values, the habits, the culture — are the things that ultimately determine whether a company will succeed. The Philips story is, in the end, a story about culture.[1, 2, 6]

Comparison: Philips and its semiconductor children (2024)

Company	Headquarters	Founded	Philips link	Market cap (2024)	Focus
Philips	Amsterdam, Netherlands	1891	(the parent)	~€25 billion	Healthcare, personal health, lighting
ASML	Veldhoven, Netherlands	1984 (spin-out)	Direct: founded as Philips/ASM JV	~€350 billion	Lithography (monopoly on EUV)
TSMC	Hsinchu, Taiwan	1987 (ITRI spin-out)	Indirect: built on Philips-ITRI CMOS tech transfer (1983)	~$800 billion	Foundry (manufactures the majority of advanced chips)
NXP Semiconductors	Eindhoven, Netherlands	2006 (Philips semi spin-off)	Direct: bought Philips semiconductor group	~$60 billion	Automotive, IoT, secure connectivity
Signify	Eindhoven, Netherlands	2016 (spin-out)	Direct: Philips lighting	~€3 billion	LED lighting, connected lighting

Key numbers

• 1891: year of Philips’s founding in Eindhoven.




• 1914: year the Philips NatLab was established.




• 1961: year Signetics (the Philips/Fairchild joint venture) was founded, commercialising the 7400 TTL family.




• 1983: year of the Philips-ITRI CMOS technology transfer agreement.




• 1984: year of ASML’s founding as a Philips/ASM International joint venture.




• 1987: year of TSMC’s founding by Morris Chang, building on the Philips-ITRI pilot line.
• 2006: year Philips sold its semiconductor group to KKR (forming NXP Semiconductors).




• 2016: year Philips spun off its lighting business as Signify.




• ~2,000: peak number of researchers at Philips NatLab.




• ~€350 billion: market capitalisation of ASML (2024).




• ~$800 billion: market capitalisation of TSMC (2024).




• ~€25 billion: market capitalisation of Philips (2024).




• ~14x: ratio of TSMC’s market cap to Philips’s (2024).

References

1. Wikipedia, “Philips.” (Standard reference on Philips, with sections on the founding and the semiconductor business.) https://en.wikipedia.org/wiki/Philips
2. Wikipedia, “ASML.” (Standard reference on ASML, with a section on the Philips founding.) https://en.wikipedia.org/wiki/ASML
3. Wikipedia, “ASM International.” (Standard reference on ASM International, the Dutch partner in the ASML joint venture.) https://en.wikipedia.org/wiki/ASM_International
4. Wikipedia, “Philips NatLab.” (Standard reference on the Philips Natuurkundig Laboratorium, the research lab behind the CD and many other innovations.) https://en.wikipedia.org/wiki/Philips_NatLab
5. Wikipedia, “Signetics.” (Standard reference on Signetics, the 1961 Philips/Fairchild joint venture that commercialised the 7400 TTL family.) https://en.wikipedia.org/wiki/Signetics
6. Wikipedia, “TSMC.” (Standard reference on TSMC, with a section on the founding and the Philips-ITRI technology transfer.) https://en.wikipedia.org/wiki/TSMC
7. Wikipedia, “Morris Chang.” (Standard reference on Morris Chang, founder of TSMC, with a section on the ITRI years.) https://en.wikipedia.org/wiki/Morris_Chang
8. Wikipedia, “Industrial Technology Research Institute.” (Standard reference on ITRI, the Taiwanese R&D institute that founded UMC and TSMC.) https://en.wikipedia.org/wiki/Industrial_Technology_Research_Institute
9. Wikipedia, “NXP Semiconductors.” (Standard reference on NXP, the 2006 buyer of Philips’s semiconductor group.) https://en.wikipedia.org/wiki/NXP_Semiconductors
10. Wikipedia, “Signify.” (Standard reference on Signify, the 2016 spin-off of Philips Lighting.) https://en.wikipedia.org/wiki/Signify
11. Wikipedia, “Wisse Dekker.” (Standard reference on Wisse Dekker, the Philips CEO who oversaw the 1984 ASML spin-out.) https://en.wikipedia.org/wiki/Wisse_Dekker
12. Wikipedia, “Gerard Philips.” (Standard reference on Gerard Philips, co-founder of the company.) https://en.wikipedia.org/wiki/Gerard_Philips
13. Wikipedia, “Anton Philips.” (Standard reference on Anton Philips, Gerard’s brother who turned the company into Europe’s largest light-bulb maker.) https://en.wikipedia.org/wiki/Anton_Philips
14. Wikipedia, “Frederik Philips.” (Standard reference on Frederik Philips, Gerard’s father and the original founder of the Philips company.) https://en.wikipedia.org/wiki/Frederik_Philips
15. Wikipedia, “7400-series integrated circuits.” (Standard reference on the 7400 TTL family, the first commercial product of Signetics.) https://en.wikipedia.org/wiki/7400-series_integrated_circuits
16. Wikipedia, “Wisse Dekker.” (Standard reference on Wisse Dekker, the Philips CEO who oversaw the lithography spin-out.) https://en.wikipedia.org/wiki/Wisse_Dekker
17. Wikipedia, “PAS 2000.” (Standard reference on ASML’s first stepper, the PAS 2000, delivered in 1986.) https://en.wikipedia.org/wiki/ASM_Lithography
18. Wikipedia, “Hsinchu Science Park.” (Standard reference on the Hsinchu Science Park, the campus that hosts both ITRI and TSMC.) https://en.wikipedia.org/wiki/Hsinchu_Science_Park
19. Wikipedia, “United Microelectronics Corporation.” (Standard reference on UMC, the 1980 ITRI spin-out that preceded TSMC.) https://en.wikipedia.org/wiki/United_Microelectronics_Corporation
20. Wikipedia, “Compact disc.” (Standard reference on the compact disc, developed by Philips and Sony in the late 1970s.) https://en.wikipedia.org/wiki/Compact_disc