Not only are the cellphone chipmakers facing potential pressure from economic uncertainty and consolidation among their customers, but the traditional players are also battling increasingly powerful challenges from Asian vendors, notably MediaTek. Among those in the most precarious positions are the Europeans ‘ NXP of the Netherlands, Swiss-based STMicro, and Germany’s Infineon ‘ which have to support very broad product ranges including automotive, industrial, consumer electronics and mobile; have huge amounts of legacy business to support; and can rarely be regarded as agile or cost competitive.
All have been seeking to streamline their product ranges and their value chains and so become more viable to bid for deals with major customers like Nokia, but they may be forced to go a step further and pool their resources altogether. A former STMicro executive has submitted a proposal to this effect to potential partners and to French president Nicolas Sarkozy, but while a joint venture would have a major impact on the structure of the handset supply chain (and other industries), extending a wave of consolidation among smaller players up to the top tier, it might create a totally unwieldy monster rather than generating the economies of scale and end-to-end technology range the market is looking for.
Joseph Borel, former executive VP of central R&D at STMicro, has proposed that NXP, ST and Infineon should consolidate into one super-chipmaker. Borel, now at the R&D consultancy JB-R&D, told electronics newspaper EETimes Europe that the trio should join to achieve critical mass that would approach Intel’s and surpass Samsung’s, spreading across a range of semiconductor sectors. He pointed out that Intel’s revenues are equal to those of Samsung plus TI, and ahead of those of the three Europeans put together ($31.3 billion for Intel versus $26.5 billion for the Euro-trio in 2006).
According to Borel, in fiscal year 2007, STMicro reported a net loss of $477 million on revenues of $10 billion; NXP made a net loss of $750 million on revenue of $7.04 billion; and Infineon, excluding its troubled DRAM subsidiary Qimonda, announced a profit of about $121 million on revenues of $6.27 billion. Against that uninviting backdrop, process R&D, design and manufacturing costs are rising rapidly as the manufacturers move to 45nm and below, and these costs can only be profitably borne by an Intel-scale giant with major market share worldwide.
“The only way to save Europe is to be just behind Intel and to put everyone under the same banner, so as to avoid duplication,” stated Borel. “We need to have the same ambition as Intel, and, with a unique management, we would be able to develop a solution that would equal Intel’s Merom processor’ [Intel’s premier line of notebook processors]. Borel added, in his EETimes interview, that a combined European powerhouse would have close ties to EU-based system houses, such as Nokia and Bosch, and more formalized links with large European research centers, such as IMEC and Fraunhofer. The three would be able to share process technologies, libraries, design developments and production facilities.
“The time of the Gauls or of the Romans is over. We should all start working together and establish a European champion,” he said.
Whatever the attractions of such a plan, from the point of view of keeping the European chip industry competitive, in reality it is hard to imagine a three-way merger of such scale, between highly political companies with national governments as well as shareholders taking an interest.
Infineon was the only one of the three to respond, reminding the market of statements made earlier this year by CEO Wolfgang Ziebart, to the effect that size is no longer the decisive factor in the semiconductor market. “The major thing giving semiconductor makers a competitive advantage has evaporated,” he said in an interview then. “The consolidation process in the semiconductor industry will not lead to the emergence of a few, large enterprises as has been stated frequently. Today, synergies result mainly through R&D and not through manufacturing, as was the case in earlier times.”