By: Phil Kernen

The supply chain bottlenecks experienced over the last two years originated with component shortages, the blame for which stemmed from insufficient semiconductors or chips. The lack of chips has denied car makers billions in sales and delayed the adoption of phones and other connected devices that use 5G networks. What is causing the problems, what solutions are being pursued, and are there opportunities for investors?

The problem

We often refer to the semiconductor industry as a singular entity. Yet, it includes hundreds of companies broken into industry groups, including designers, suppliers, manufacturers, vendors, and distributors—many of which operate in more than one. Semiconductors categories are many and varied: memory, logic, analog, discrete, optical, and sensors. Most manufacturers focus on select elements, but end users typically require products from all segments and rely on multiple suppliers. The absence of a single specialized chip can bring production to a halt even if all other components are available.

The primary chokepoint stems from wafer size. Wafers are thin slices of semiconductors on which manufacturers fabricate integrated circuits. Chipmakers buy wafers to process them into chips. Over the years, chipmakers migrated to progressively larger wafer sizes measured by diameter and width. Each advance, assuming high plant utilization, allowed the production of 2.2x more circuitry per wafer and reduced manufacturing costs and cost-effectiveness. 

Like a car factory set up to produce a specific model, fabrication plants work with a single wafer size, and many chipmakers use 300mm wafers, which have been around for two decades. Over the next two years, forecasts call for 300mm wafer supply to fall short of demand by 10%. Bigger wafers cost more to produce, and most of the biggest wafer producers are spending billions on new facilities that won’t be in operation until 2024. 

But 200mm wafers, which first arrived in the 90s, didn’t disappear. When balancing cost and efficiency, bigger isn’t always better, and not every chip type migrates to 300mm. Many devices built from 200mm wafers are long-lived products and will stay on existing technologies for some time. Other new applications want to take advantage of the technology/cost benefits of 200mm wafers or feel migrating to 300mm isn’t worth the cost. As a result, chips built on 200mm wafers remain on the critical path to meeting the growing demand for applications like 5G, automotive, and the Internet of Things (IoT). While billions of dollars are flowing into new 300mm wafer production capacity, far less is earmarked for additional 200mm capacity. It is like a bike manufacturer who can obtain the latest gear assemblies but can’t get enough basics like wheel bearings or pedals.


To help address the supply constraint, corporations and governments are making record-breaking investments in chip manufacturing plants for economic and political reasons. Worldwide projections call for $109 billion in investment in fabrication facilities and equipment in 2022, double the level of 2019. Forecasts don’t delineate between wafer sizes. From a geographic standing, equipment spending is led by Taiwan, Korea, and China, with the Americas in fourth place.

The U.S. has the highest presence of chip designers and the manufacture of chipmaking equipment, Taiwan has the highest concentration of fabrication facilities, and Japan represents the largest share of wafer manufacturing. No local market has all the capabilities required for end-to-end semiconductor design and manufacturing, but this may change. 

Regarding resources devoted to semiconductor productivity, China leads them all with a $170 billion pool dedicated to funding companies and building 31 major semiconductor factories over the four years through 2024. That total exceeds the 19 expected in Taiwan and 12 scheduled in the U.S. Most Chinese factories will make chips containing older, more mature technologies rather than the cutting-edge processors focused on elsewhere (think 200mm vs. 300mm). Industry observers expect China will be making more advanced chips before long.  

The Chinese push for semiconductor self-sufficiency explains why the E.U. is debating a bill to offer industry subsidies for $40+ billion, and the U.S. is about to pass a $52 billion legislative package of its own. The tenuous relationship between China and Taiwan, and the considerable concentration of manufacturing capacity in Taiwan, underscores the geopolitical risks of broken semiconductor supply chains.  


Semiconductors have always displayed an element of boom or bust. Design cycles and extended fabrication lead times can quickly swing the supply chain from deficit to surplus. Projecting consumer tastes and demand years in advance is difficult. Expected government-driven capital injections for additional capacity complicate it further, cloaking a business decision in a political wrapper. However, our collective demand for new gadgets, electric automobiles, mobile devices and the chips that power them suggests demand will continue to grow. Picking the best operators in each segment should provide good, long-term returns for investors willing to do the research.

Disclosure: This is for informational purposes only and any reference to a specific company (or industry) does not constitute a recommendation to buy or sell that company (or a company in that industry). The reader should not assume that an investment in a security in the industry identified or described, was or will, be profitable.   

 For a complete list of disclosures, please click