The high investment required for mass chip production and the intensity and cost of R&D in an extremely competitive market constitute formidable barriers for potential new entrants, particularly those from developing countries.
While the production of integrated circuits is beyond the reach of most developing countries, the design, particularly of custom and semi-custom chips, has been undertaken in many of them, particularly in the newly industrializing countries (NICs). This has been facilitated, on the one side, by the development of computer aided design tools that can run on relatively small-sized computers; and on the other, by the possibility of contracting with various silicon foundries the manufacture of a chip according to independently made designs.
Technological advance in the semiconductor industry is an interactive, cumulative process where improvements are directly based on the preexisting stock of knowledge. Studies on the role of IPRs [Intellectual Property Rights] in promoting innovation in this industry have shown that gaining lead-time and exploiting learning-curve advantages are the primary methods for appropriating the return on investments in R&D (Levin et al, 1987, p.788).
The protection of layout designs of integrated circuits as a specific
subject matter was born in the United States in 1984, with the approval
of the Semiconductor Chip Protection Act (SCPA). The
growing concern over the decline of the United States' competitive advantages
in chip production and trade during the 1980s prompted Congress to adopt
a
The
The SCPA, in addition, included stringent reciprocity rules that forced Japan to adopt similar legislation ""Act concerning the circuit lay-out of a semiconductor integrated circuit" (law No. 43)), followed by the European Communities (Council Directive on the legal protection of topographies of semiconductor products, 87/54/EEC).
The World Intellectual Property Organization, shortly after the enactment
of the SCPA, initiated studies and consultations in order to establish
an international treaty of the matter, based on the
These perceived shortcomings were expeditiously addressed a few years later. The TRIPS Agreement practically derogated the Treaty's provisions that had been rejected by the two chip powers, and added those obligations that they felt were missing.
So far, only some developing countries have considered legislation to specifically protect integrated cicuits' designs, but all WTO Member countries are obliged to do so.
One of the main reasons for the enactment of the SCPA and its internationalization was the alleged copying of original chip layout designs, particularly by Japanese competitors. However, the very few instances of litigation that took place on the basis of that regime would indicate that the copying of chips' layout design was not the main bone of contention[1].
The main purpose of the
The important controversies did not relate to the layour designs, but to the technical ideas underlying them or to their production processes subject to patent protection.
Thus, Texas Instruments was reported to have earned (by 1994) more than US$1.5 billion in royalties from its patent portfolio. Its main source of income was a broad patent originally filed 30 years ago "covering an old-fashioned technology that has been extended by peculiarities unique to the US patent system". Intel also has an aggressive IPRs strategy with a litigation budget of at least US$100 million, that few companies can match. Based on its patent US4,338,675, it has attempted to bar competition not only from other major chip producers but also from small design houses (Warshofsky, 1994, pp. 252 and 256).
Action by Texas Instruments reached competitors worldwide. It sued on the same charges at the same time seven Japanese companies, which have reportedly paid an average of around US$30 million per firm. Samsung was also sued, but it paid over US$90 million. The reason for this different outcome seems to be that the Japanese firms held "several patents of their own and were able to negotiate cross licensing agreements and thus lower the amount of royalty payments to Texas Instruments (Moody, 1989, p.38).
While the SCPA form of intellectual property protection from the semiconductor industry has been widely panned by engineers as ineffective, there are definite areas in which this legislation can have dramatic positive benefits to the overall protection afforded an IC, especially considering the relatively low cost of registration. As evidenced by the graph in Figure 6 [ which shows world registrations running at around 10000 pa from 1985-1994] registration of mask works has been relatively constant since the SCPA inception in 1985. This fact gives credence to the theory that while engineers may grouse at the incomplete nature of the SCPA's protection, they nonetheless consider it a viable IP protection tool in the semiconductor era.
Again, one must consider the IP protection afforded an IC as a 'mesh' of
individual components which together provide a framework of protection greater
than each individual protective element alone. Given the SCPA's track record
inside and outside the courts, one must conclude that there is still life
in this legislation and it does indeed have real application in everyday
semiconductor IP practice." Kevin Klughart,