5G – Who’s Right?
The Nikkei Asian Review has stated that “China’s efforts to build 5th generation wireless communications infrastructure have lost momentum because equipment manufacturers have used up American-made parts and components, forcing relevant merchant suppliers to turn to the US and European markets.” The article quoted Murata (6981.JP) chairman saying that Chinese companies’ demand for parts to prevent network interruptions has declined and another communications component manufacturer stating that since the summer of 2020 it is no longer supplying parts to Huawei.
According to the Chinese Ministry of Industry Information Technology the current number of 5G base stations in China (10/19/21) is 1.159m but the joint resolution of 10 departments focused on 5G development have set a goal of 18 base stations per 10,000 people by 2023, which would target the number of base stations at that time to be 2.52m based on a 1.4b population estimate. In the first year of implementation (actually ½ year) 130,000 base stations were put into service. Last year 600,000 were added, at a speed that was about 2x the 2019 rate and this year’s target for new 5G base station adds is 1m, however based on the MIIT numbers, the total installations for this year will be ~500,000 for a total of 1.23m units at year end.. This would imply 645,000 new 5G base station adds in both 2022 and 2023 to meet China’s goal, but does prove out that construction of new base stations has slowed from 2020 levels.
The bigger question is why has the rate of new 5G base station installation slowed? Is it because Huawei and ZTE have run out of chip inventory and are unable to procure new silicon from suppliers who fear producing same would put them in the sights of US trade officials? More than likely this is not the case, and while Huawei and ZTE have certainly been affected by the US trade sanctions in their smartphone businesses, early on Huawei stated that their biggest priority (this is all before the final portion of the trade sanctions were put in place) was making sure it had a stockpile of the silicon needed for its 5G base station products and began building such inventory through massive purchases from Taiwan Semiconductor (TSM) as far back as 4Q 2019. Estimates that Huawei’s orders from TSM before the final sanctions were more than 2m units and Huawei has stated that its base station component inventory was sufficient to cover construction through this year and beyond and had eliminated other US components from their base station products early in 2020.
It all comes down to who you believe. Japan (mostly Nikkei Asian Review) continues to tout the idea that Hauwei and ZTE are out of components, causing a slowdown in 5G roll-outs, while China says 2020 was the ‘big push’ year and is still on track to meet its goal. Huawei does have the option of using locally produced communication silicon from SMIC (688981.CH) or other Chinese suppliers, which would be based on the same chip design developed by Huawei affiliate HiSilicon (pvt), although NAR says this could be inferior to TSM components. Given that there is little chance we would be able to see and identify BOM sources for Huawei base stations placed on the Mainland, we have no way of confirming that Huawei is still using TSM silicon, nor do we take NAR’s view that blanket statements by Japanese suppliers hold weight, so it seems we will have to wait until the end of 2Q 2022 to see if MITT’s data indicates that China is still on track to meet its goal. In the interim we watch peripheral data sources for any clues as to where Huawei’s base station inventory lies and whether the implementation slowdown is forced or in line with plan.