The Hidden Cost Of Tariffs

The impact of tariffs on the semiconductor industry is just now being assessed, but there’s a lot more to this picture than import and export duties.

In fact, the biggest and longest-lasting effects may have less to do with taxing imports than what happens across the global supply chain that includes everything from manufacturing equipment to materials to investment capital.

Import duties are a political reaction to an amalgam of political and social issues. Every country has used them at one time or another, often with unexpected results. The promise is that by raising prices on imports, more people will buy locally sourced products because they’re less expensive than the imported products. And, at least in theory, businesses within a country should be better off.

Last week, the U.S. Trade Representative announced a 25% tariff on $16 billion worth of imports from China. SEMI estimates it will cost its 400 U.S.-based companies more than $500 million annually.

But when it comes to semiconductor manufacturing, this formula cannot be applied the same way as it can for raw materials such as steel, aluminum or soybeans. There are three factors that need to be considered:

Raw materials. Just as with most technology, the number of components and raw materials developed in other countries represents a substantial portion of the overall value in those devices. How will the prices of hafnium, tungsten and copper be affected?

Rare earths are a particular worry because they are no longer mined in the United States. The last U.S.-based mine, located in Mountain Pass, California, shut down in 2015 when the owner, Molycorp, filed for bankruptcy. The company said it could no longer compete with global pricing, which had fallen through the floor. Since then, buyers from around the world have attempted to purchase the mine, only to be stopped by the Committee on Foreign Investment in the U.S., aka CFIUS.

It’s not clear how expensive it would be to get the mine up and running, or how high rare earths prices would have to rise to make that mine profitable. Rare earths are used in cell phone and electric vehicle batteries, magnets, lasers and to boost the strength of alloys.

Manufacturing capacity. While much of the attention on chip manufacturing has centered around solving the lithography at 7/5/3nm, most of the volume will continue to be at older nodes using 200mm wafers.

It’s no secret there is a shortage of 200mm equipment and capacity. What is less obvious is that much of the new capacity being added to address that shortage is in China. There are eight new 200mm fabs under construction in China, which will be used to make chips for various segments of the IoT, IIoT, sensors, MEMs chips, and a variety of other functions that may sit next to a 7/5/3nm ASIC in a smart phone or a cloud-based server rack.

On the flip side of this, much of the equipment developed for 200mm fabs is manufactured in the United States. For several years, there was huge demand for used equipment because it was cheaper, but most of the equipment that could be refurbished has already been purchased, and there are still new fabs to fill.

All of this could raise the prices of both 200mm and 300mm chips, because tariffs may be imposed on both sides. In a complex supply chain, it’s possible that companies may be paying tariffs on multiple sides of a trade war.

Money. Possibly the least understood impact because it is often private is investment capital. China’s semiconductor investment fund at last count was somewhere in the neighborhood of $300 billion. That money initially was targeted for acquisitions, but more recently it has gone into funding startups inside of China.

Normally, it takes five to seven years for startups to begin paying back investments, and it can take up to a decade with very complicated technology. What the impact of this will be is hard to assess in the early stages, but it will become more noticeable as these companies mature. It’s hard to put a price tag on the free flow of information in a startup market, but a trade war is bad for business and that information exchange.

Bottom line. A trade war in a globally integrated industry will raise prices on multiple levels, and with increasingly complex technology it will take years before alternatives can be developed.