For the automotive and semiconductor industries, the past few years have been a journey of extremes. In both cases, the pandemic precipitated a devastating drop in demand. Then they both saw a big rebound as consumers sought to get back on the road and also increased the use of computers and other devices for remote work. And now, again, the two sectors face a similar situation, albeit from different ends of the spectrum: dealing with the chronic shortage of chips.
Even with fabs operating flat out, chipmakers have not been able to meet demand, and this situation looks likely to last well into 2023. Boosting production is the obvious answer—but not a simple one. The semiconductor industry has increased its production capacity by nearly 180% since 2000; even so, capacity is nearly exhausted. Building new plants takes many years and many billions in investment.
The situation has not improved over the past year. Indeed, the war in Ukraine has made things worse. Ukraine accounts for 25 to 35% of purified Neon gas production, and Russia supplies 35% of palladium, an essential input.
To put it bluntly, the auto industry cannot expect supply to meet demand any time soon
Automakers need no reminding of this. Chip shortages have forced more than a few to halt production, as Toyota recently announced. Others have cut back on options. Overall, there were 1.7 million fewer vehicles built in 2021 than in 2019. A few carmakers were able to safeguard profits with manufacturing and sales strategies designed to optimise margins. But that raises problems, too, such as a shortage of lower-margin vehicles and more fluctuations in chip demand.
All semiconductor-enabled industries, such as consumer electronics and wireless technologies, face this problem. But the situation may be more acute for the auto industry. Carmakers and suppliers typically follow a just-in-time manufacturing strategy. Since many players didn’t expect the chip shortage in 2020 and 2021, they had limited stock available, which disrupted the entire supply chain. On the demand side, vehicles are complex and personalised products, making it more difficult to predict needs. Finally, the cars of the future will need more chips, and more sophisticated ones, as electrification, advanced driver assistance systems, and connected features become more popular.
In response to these pressures, some carmakers have started to order more chips than they need. Others are requesting “take or pay” contracts in which they can either accept a certain quantity of chips or pay a fee if they decline to do so; this arrangement helps companies match chip demand to manufacturing capacity. But these actions fall short of being fixes. Building up inventory, for example, can work for a single company, but exacerbates the overall shortage. Changing suppliers is also unlikely to help much, because it can take six months to a year; chipmakers have to go through a complex qualification process, including meeting specific intellectual property requirements.
To put it bluntly, the auto industry cannot expect supply to meet demand any time soon. They therefore need to build a foundation for long-term strategic supply management. Some are taking steps in this direction, for example by setting up war rooms dedicated to knowing as much as possible about the dynamics of the market and the competitive landscape. They have developed dashboards that automatically generate and combine data from a wide range of sources to reveal insights on a company’s supply chain and a semiconductor player’s commitments. The goal is to provide accurate data to inform decision-making, including what technologies to prioritise.
Other short-term options include offering extra payments to expedite the production of wafers when capacity amounts to less than 5% of production volume and replacing back-ordered components with similar but more feature-rich units (for example, swapping in chips with more memory).
These are all no-regrets moves. They are also only a start.
For long-term resilience, companies should create strong technology maps that define their semiconductor needs. These improve transparency with suppliers, allowing the companies to determine product development more strategically. Developing a clear-eyed view of supply uncertainties and their dependency on selected components makes for better, more informed decision-making. Carmakers could decide to re-engineer chips, for example, or to reduce variations among vehicle components.
With a roadmap in hand, they can turn to demand planning, both short- and long term. Internally, sales teams can use advanced analytics to forecast required demand, while production teams can use digital tools to help them allocate semiconductors to the right areas. These steps can help address issues that interfere with chip supply, such as a lack of insight about the semiconductor content of next-generation products.
But demand planning needs to go further. In other industries, such as electronic equipment, manufacturers and suppliers share information. This helps both parties to recognise small potential problems before they become big ones. The automotive industry has not done this in the past, because it didn’t need to. Now it does. A transparent, two-way supply-demand analysis can allow chipmakers to confirm that they are able to fill orders and to allocate chips to potential customers earlier. Building a shared long-term perspective on demand allows carmakers and chipmakers to consider investing in projects together, and thus share the financial burden while improving the supply of low-margin or innovative technologies.
Finally, auto industry players may want to rethink the way they structure contracts for semiconductor-related sourcing. They could make up-front volume commitments more binding, guaranteeing demand for up to a year for technologies common to multiple chips. They could also consider making selective investments in supply-chain resilience. Pandemic-related supply-chain shocks were dramatic, but cannot be considered unexpected. McKinsey research has estimated that companies can expect a supply-chain disruption lasting a month or more every 3.7 years, on average. Options include dual-source manufacturing qualification with chipmakers; adjusting pricing levels with supply guarantees; and bundling volumes to achieve greater negotiation power.
The key takeaway here is that the chip crisis is not going away in the next few weeks or months. Automakers and chipmakers who work together to address the imbalance in demand may feel somewhat less pain—and gain an enduring competitive advantage
About the author: Ondrej Burkacky is a Senior Partner in McKinsey & Company’s Semiconductor Practice