It’s crucial that Europe regains a significant position in advanced semiconductor manufacturing, but it would be foolish to rush into it.
Carmakers trying to rev up production after last year’s emergency stop quickly ran out of engine control units and parking sensors because their suppliers couldn’t get their hands on automotive chips anymore. After demand for car electronics plummeted during the first corona wave, the world’s foundries happily turned to chips for electronic devices such as laptops and tablets, which were – and still are – in high demand because of remote work and studying. When car sales started to recover, the automotive supply chain found itself at the back of the line.
Volkswagen was the first to sound the alarm in December, and German Economy Minister Peter Altmaier recently appealed to the Taiwanese government to boost supply for his ailing car industry. In an apparent vaccines-for-chips swap, TSMC has agreed to expedite production of automotive products by reallocating orders. Even though the Taiwanese foundry will engage in a rare Super Hot Run, it will take months before carmakers get the parts they need. The shortage could stretch into Q3, according to market research firm IHS Markit.
I don’t pity carmakers. They notoriously squeeze their suppliers dry, expecting them to be at their every beck and call. Now, they finally bumped into a source of supply they can’t bully. Apparently, that hasn’t quite sunk in: Volkswagen is considering slapping their Tier 1 suppliers with damage claims.
The shortage is, however, yet another wake-up call for Europe. Both China’s and the Trump administration’s disregard for the rules of the liberal international order already made it abundantly clear that the EU will have to fend for itself much more than it has grown accustomed to. When free trade is no longer a given, the sparsity of high-tech manufacturing capabilities in Europe reveals some very uncomfortable vulnerabilities. And semiconductors are at the center of many of them.
Test less, verify Moore
At the Bits&Chips Event 2023 on 12 October 2023, Phillipa Hopcroft from Crocotec and Ivo ter Horst from ASML will kick off with a keynote speech on transforming how lithography machine software is built. Make sure to keep an eye on the website for program updates and save the date!
And so we’ve seen all sorts of calls and initiatives pop up to boost semiconductor manufacturing on European soil. The most important one being a new European Commission’s Important Project of Common European Interest (IPCEI) for microelectronics, in which state-aid and funding rules are relaxed to address market failures in strategically important areas. Altmaier expects the current one to trigger investments of up to 50 billion euros (this number applies to Europe as a whole, not just Germany).
Glorified assembly lines
It should certainly be possible for an alliance of the likes of Volkswagen, Infineon, Bosch and Globalfoundries to successfully reduce their dependence on Asian foundries for automotive electronics, which are generally manufactured using mature process nodes. That will keep German engines purring for a couple more years.
Cars, however, are rapidly evolving into intelligent electronic devices on wheels. As more and more sophisticated driver assistance systems and, eventually, self-driving abilities are installed, increasingly leading-edge chips will slip into cars. Meanwhile, the combustion engine will become a thing of the past. This has vast consequences for the car industry. The added value will shift to electronics, ie to Asia. Without their own solutions, German car plants will devolve into glorified assembly lines.
This is just one example of why it’s crucial that Europe regains a significant position in advanced semiconductors, including manufacturing. Without it, our industries risk being deprived of the oxygen that they need to generate wealth in the future.
The fact that Europe really needs to catch up appears to have sunk in. Nineteen EU member states recently signed a joint declaration to collaborate on “strengthening Europe’s capacity to develop the next generation of processors and semiconductors. This includes chips and embedded systems that offer the best performance for specific applications across a wide range of sectors as well as leading-edge manufacturing progressively advancing towards 2nm nodes for processor technology.”
Measured steps
Unfortunately, as much as I’d like to, I don’t share Malcolm Penn’s optimism that, taking advantage of an upcoming technology transition, such a European leading-edge fab is viable in the near future. If the willingness of companies and most member states to go all-in like that isn’t already a sheer insurmountable obstacle, I doubt we can go head-to-head with TSMC and Samsung. Their head start is simply too big, and the upcoming gate-all-around technology isn’t all that different from FinFET anyway. And what chips would the fab manufacture, anyway?
It would nonetheless be unwise to abandon all ambitions and have the IPCEI become yet again a trough for the vested interests of the fragmented European technological industries to feed at without addressing the strategic needs of the region. Rather, let that leading-edge fab become a long-term ambition, our moonshot if you will, and take one step at a time to get there. Make no mistake, though, these steps will still need to be much bolder than they were in the past.
In her new book “Mission economy,” star economist Mariana Mazzucato argues yet again that it’s governments rather than businesses that need to take the lead in the development of new technologies. Their role should be to “tilt the playing field” in a direction that’s good for societies as a whole. This doesn’t mean switching to the planned economy but rather that governments become more assertive in getting the things done that societies, not impatient shareholders, need. If our industries want R&D support, let’s give it to them – but with some strings attached.
Let’s lure companies out of their comfort zones. And create environments where new ones can thrive. If Europe can’t manufacture smartphone processors, what about IoT chips? How can we secure a local supply chain for our digital infrastructure (6G and data centers)? What about the next-gen automotive ICs? Tesla designs them in-house, why can’t Volkswagen? Europe needs to gradually expand its technological prowess to a point where building that advanced fab actually makes sense.
It won’t be easy and it won’t be cheap. But as European Commissioner Neelie Kroes said a couple of years ago, when she attempted to resuscitate European semiconductor manufacturing: “Complacency is a killer. The digital economy is growing seven times faster than the rest of the economy, we need to take risks. There’s too much risk-avoiding in Brussels.”
Kroes isn’t a socialist by any stretch of the imagination, but she knew when a healthy bit of chip socialism is in order. Forget about a Great Leap Forward, let’s give a couple of measured steps a try.