Wealth Management: Strategy blog – The semiconductor shortage
Strategy team: Charlie Hines
Chips with everything
Semiconductors – also known as microchips and integrated circuits – are components found, in some form or another, in most products that use electronics, from complex supercomputers to cars, refrigerators and the smart cards in our wallet.
The semiconductor industry is making headlines today because a shortage of chips is interrupting the production lines of manufacturers using them as a key input – most notably, in the auto sector. The shortage is not in the designing of chips, or the manufacturing of the complex equipment needed to make them, but in the chips themselves. Companies that manufacture chips in industrial quantities – in fabrication plants ("FABs") that can be massive in scale – are not able to keep pace with demand, particularly at the high-performing, more complex end of the market.
Globally, the three main chipmaking manufacturers with their own fabrication plants are Taiwan Semiconductor Manufacturing Company (TSMC) – a pure foundry business, arguably the most 'bespoke' and technologically advanced, but whose sole purpose is to manufacture chips –, and Samsung and Intel – Integrated Device Manufacturers ("IDMs") that design, manufacture and then sell these chips. Other 'fabless' chipmakers such as NVIDIA, Qualcomm and Apple outsource their chip manufacturing to these few companies.
Chip making is specialised, and the contracts with these suppliers are large. The industry faces high barriers to entry, large amounts of R&D and infrastructure spending – depending on how advanced the chip is, fab plants can cost anywhere from $5bn up to about $15bn to create, and for advanced plants, machinery will be a large part of the cost. Chip making isn't especially quick either: the manufacturing process takes from months up to almost a year for the most advanced chips. The facilities involved are high-tech as well, with production rooms reportedly kept cleaner than hospital operating theatres – a single particle of dust within the production line can cause major disruption.
The semiconductor market continues to advance at pace. Chips are getting smaller and the number of transistors each chip can hold is continuing to increase – the number of transistors per unit area doubles about every two years - making them more energy efficient, faster, and more powerful. There is however a physical limitation to this, and the incremental efficiency gains will eventually start to slow. For reference, 28.3bn transistors can today be packed on a chip with an area of 6.28cm², with each measuring 5 billionths of a meter. For comparison, in 1971 when Intel released their first microprocessor, 2,300 transistors were packed onto a chip, with each transistor measuring 10 millionths of a meter.
Why is there a shortage?
Semiconductors are in focus because their scarcity is restricting auto production in particular. The nature of the shortage is however a little more subtle than it appears. Last year's covid-led shutdowns, and uncertainty around the speed of recovery, led automakers to cancel their chip orders. Meanwhile, other sectors, such as consumer electronics, saw rapid increases in demand, and absorbed the spare capacity. As a result, automakers effectively lost their place in the queue, and given the long lead time for supplying chips, and automakers' lack of inventory, it is likely to be some time before supply issues in the auto sector are resolved – Ford, for example, has warned of a 50% reduction in output in the second quarter of this year. In the meantime, however, those other industries who moved up the queue will be seeing their output grow more in line with booming demand.
Supply issues are not limited to the auto sector, and chips' ubiquity in modern life and the burgeoning "internet of things" means that if an overall shortage of chips were to materialise it would have big implications for global output – or at least, for its growth. Meanwhile, fabrication is stretched, not shrinking, and some customers are faring better than others. Despite the high-profile shortages, overall chip sales are still projected to grow by 11% this year.
Supply shortages are not new in the industry, though previous shortages have predominately been more isolated events for specific types of chip. The supply pinch this time is not just related to one type of chip, but partly a reflection of one big customer's (the auto sector) ordering and inventory misjudgements (that said, if the auto sector had secured all the chips it suddenly needed, then those other customers would presumably have lost out). Location specific issues – the Texan freeze earlier this year, and a fire at a Japanese manufacturing plant – have not helped either.
Semiconductor supply is not the whole story – demand is increasing rapidly in the short-term as economies reopen, and more steadily in the longer-term as more uses for chips emerge, and products now require more chips to function. Taiwan Semiconductor Manufacturing Company is due to spend $100bn over the next three years to boost capacity, but high barriers to entry for new manufacturers and the overreliance on a few large fabricators suggest a structural element to the issue.
There is a political angle too. Trump's export ban last year relating to the US supply of semiconductors and chipmaking equipment has highlighted the potentially fragile nature of global supply chains. Countries are increasingly looking for greater self-sufficiency, including the US itself: Biden's latest infrastructure package however has earmarked roughly $50bn in spending for domestic production of semiconductors. Likewise, China's latest 5-year plan has pledged to make chip production a national priority.
What does this mean for investors? First, at the macro level, it suggests that some of today's bottlenecks and pinchpoints may not disappear overnight. We've noted that the April spike in the US CPI is unlikely to be the main inflation event – but to the extent that semiconductor constraints are structural, and other big users start to face constraints, that main event may not be quite so far behind. Second, and more specifically, while the immediate impact of disruption seems to have rebounded against the fabricators – it is hard to be sure, because two of them are in a region that has faced a wider local setback – the healthy structural demand for their products, and their dominant position, seems to suggest some underlying competitive advantage.
Past performance is not a guide to future performance and nothing in this blog constitutes advice. Although the information and data herein are obtained from sources believed to be reliable, no representation or warranty, expressed or implied, is or will be made and, save in the case of fraud, no responsibility or liability is or will be accepted by Rothschild & Co Wealth Management UK Limited as to or in relation to the fairness, accuracy or completeness of this document or the information forming the basis of this document or for any reliance placed on this document by any person whatsoever. In particular, no representation or warranty is given as to the achievement or reasonableness of any future projections, targets, estimates or forecasts contained in this document. Furthermore, all opinions and data used in this document are subject to change without prior notice.