It may seem as if the chip shortage is over. After an extended period, starting during the pandemic, reports seem to indicate that now there is actually too much supply. Do we really have too many chips, and is this good news?
As it turns out, the situation is complicated.
It pays to understand what caused the shortage in the first place. With the pandemic and associated lockdowns, many people bought products like laptops, gaming consoles, and other electronics, causing demand to skyrocket. This occurred against a backdrop of already increasing demand due to trends like the increasing digitization of automobiles, which now contain thousands of chips and millions of lines of code. Increased use of “smart” appliances and the internet of things (IoT), cryptocurrency mining, and quantum computing, all have contributed to the upward trend of chip demand over recent years.
Meanwhile, disruptions such as the war in Ukraine and strict pandemic lockdowns in China have played a role in constraining supply. With demand outpacing supply, shortages resulted, causing major troubles across industries that rely heavily upon chips. For example, the automotive industry suffered billions of dollars in lost revenue as automakers were forced to idle their plans and produce tens of millions less vehicles because they couldn’t get the chips they needed.
But now reports show that the shortage is easing. This is largely because demand has decreased, driven by declines in the overall macroeconomic environment. With persistently high inflation for staples like food, gasoline and energy, demand for electronics is down. Consumer behavior may be changing as well. With the pandemic ending, consumers are returning to purchasing more services than goods. As a result, the third quarter of 2022 saw a 6.3% decline in global semiconductor sales from the previous quarter. In fact, many companies now have an oversupply of chips instead of a shortage. As manufacturers tried to ramp up production, the buyers of chips over-ordered and created stockpiles. These stockpiles, coupled with softening demand, has left many sitting on excess inventory.
So, is the chip shortage really over? Not exactly. Chips are not a homogenous commodity that can be readily substituted for one another. There are many different types of chips that perform different specialized functions, and particular industries have unique needs. Global averages tend to obscure these subtler dynamics. Whether the shortage is easing or persisting depends on the type of chip and the industry. For example, while the automotive industry was hit hard, the aviation sector has emerged from the shortage relatively unscathed.
One way to think about chips is to split them into broad categories based on how technologically cutting-edge they are. The most technologically advanced chips have seen slowing demand driven by a decrease in personal computer sales. Memory chips are also seeing a steep decline in demand, leading to oversupplies. On the other hand, high-end chips are still in demand for applications like iPhones and data centers. Then on the other hand, less technologically advanced, or “legacy,” chips are still experiencing constraints. Such chips are used widely in the automotive industry, where simply migrating to more advanced chips isn’t practical due to high costs of redesign. Here the shortage persists, with CEOs of Ford and Stellantis predicting the shortage to continue into 2023.
What does the future of the chip supply chain hold? Some predict that increasing digitization will increase demand over the long term. But time will tell whether the supply will catch back up. The recent passage of the CHIPS and Science Act has prompted many manufacturers to begin construction of new facilities in the US, but it will take time for new capacity to be brought online. Meanwhile, new export controls on China ban the export of US chips and chip-making equipment and software, as well as places restrictions on US citizens working for Chinese chip companies. Another factor is mounting tensions between China and Taiwan, which could have dramatic consequences for the semiconductor industry.
How can industry learn from this episode and prevent such supply/demand imbalances from happening again in the future? One area is planning. Leveraging analytics across the supply chain and AI-enable forecasting capabilities can support decision making in the short and long term. Building stronger supply chain relationships between equipment manufacturers and suppliers can facilitate data sharing and increase visibility across the supply chain. Others have proposed a strategic chip reserve in case demand once again outpaces supply. Supply chain practices such as enabling end-to-end traceability, diversifying sources of supply, and optimizing safety stock can also help to manage risk. Of course, nobody has a crystal ball to see perfectly into the future, but proactive and data-informed business planning can help to prevent shortfalls and build resilience into your supply chain.
Zachary A. Collier, Ph.D., is assistant professor of management at Radford University, and is a visiting scholar at the Center for Hardware and Embedded Systems Security Trust.