An Analysis of the Economics of Photomasks in Semciconductor Manufacturing

This paper provides survey data for benchmarking and analyzes the photomask manufacturing lifecycle. The contents of the paper can be summarized as follows: Survey data indicate that: 1) There is wide variation in mask shops in average labor content per mask, in the staffing strategies employed, and in the fraction of engineering plates produced. 2) Both scheduled and unscheduled downtime of pattern generation tools, in particular those that run above 50 kV, was exceedingly high. 3) The wide variation in the financial indicators suggests that all participants may have significant cost-reduction opportunities within their operations. To identify these opportunities, the authors propose detailed follow-on studies of cost structure, management structure, and procedures within mask shops. 4) The empirically grounded model of the photomask lifecycle suggests that the number of masks demanded, their price, and the investment patterns of photomask shops vary dramatically over the photomask manufacturing lifecycle. 5) Yield, capital productivity, and economies of scale are the critical drivers of profitability in photomask manufacturing. 6) These findings imply that a bifurcation of Moore\´s Law is likely to occur. The ASIC and foundry business will continue in 130 nm. Producing at 65 nm geometries and beyond will be reserved for high performance, high volume "best-selling" designs. Nonetheless, some of the larger foundries are pursuing 90 nm and 65 nm technology, probably in anticipation of producing some high volume best sellers. 7) The trend to outsource mask making is reversing towards vertical integration. The trend towards introducing fewer designs is likely to continue. Mask shops will not be able to take advantage of economies of scale. The resulting consolidation pressure is likely to trigger further merger-and-acquisition activity among photomask manufacturers. 8) To survive in this intensely competitive environment, photomask manufacturers must coordinate develop- ment with their customers and suppliers, as well as operate under business models that stress capital productivity