I am recommending a short in Yageo, a Taiwan-based manufacturer of multi-layer ceramic capacitors.
MLCCs are small, low-cost electrical components that store energy electrostatically and are used in a variety of consumer electronic, automotive, and industrial applications.
The MLCC market is dominated by a few large scale manufacturers primarily based in East Asia: Murata in Japan (40% share), Samsung Electronics in Korea (21% share), Yageo (6%), Taiyu Yuden (12%), and TDK (8%).
Within this industry split, Murata and Samsung have historically focused on the highest-grade, highest-end, specialized MLCC types, categories experiencing longer-term volume growth due to growing content in a wide variety of end market applications (5G, auto electrification, AI, IoT and high-end smart phones), while Yageo and other Taiwan-based manufacturers primarily produce commodity general-use MLCCs typically found in a variety of legacy industrial and consumer electronic applications.
Like most chip and component products, MLCCs have historically experienced price-downs of ~1-5% each year, but in late 2017 and throughout 2018 a perfect storm led to an acute shortage of commodity MLCCs that resulted in a dramatic increase in price and delivery lead times.
Murata, one of the key global producers of MLCCs, decided to remove significant commodity MLCC supply from the market, shuttering production capacity focused on commodity MLCC product types and rededicating production to higher-end MLCCs to align with longer-term growth. At the same time, demand for MLCCs surged due to strong global synchronized growth, with especially strong sales in auto and consumer electronic end markets. Prices of the MLCCs began to rise dramatically, leading customers to scramble and exacerbate the shortages by double-ordering and building buffer inventory to mitigate supply chain disruptions. Producers were slow to add significant supply to respond to the market shock, as lead times for new production capacity took 12-18 months to bring online and producers were wary of capacity additions after being burnt in previous boom-bust cycles. As a result of both demand shocks and supply squeezes, the prices of key MLCCs doubled over the course of 2018.
Yageo was the primary beneficiary of the MLCC shortage and my industry checks indicate they were the most aggressive on pushing through price hikes and capitalizing on the shortage (while also leading to a lot of negative goodwill from customers in the process). As a result, Yageo's revenues between 2016 and 2018 more than doubled from NT$28bn to NT$77bn in 2018 (driven by a ~2x in ASP and ~60% increase in production volumes), EBIT went from NT$4.5bn to NT$42bn, and EPS went from NT$7 to NT$85. This was clearly an unsustainable surge. I believe the MLCC cycle is peaking out for a few reasons:
CapEx increased by 30-50% by all the major producers over the course of 2018. Yageo indicated it is adding 10bn / month of MLCC capacity in 2019 and 2020. Walsin Tech, another Taiwan producer, stated it is expanding MLCC capacity by 20% in 2019. Many of the major capacity additions that were initiated in late 2017 have 12-18 month lead times (takes time to secure land permits, procure production equipment, etc.) and are now poised to come online over the course of 2019
Global growth has sharply decelerated in recent months, reversing the strong synchronized demand growth of the prior period. China specifically accounts for ~25% of global capacitor demand, and the industrial data there has been terrible. Auto demand is moderating, while iPhone and PC demand has also been soft. Chip makers like Texas Instruments, who have historically operated on a similar cycle to the passive component suppliers, have all given commentary pointing to a deteriorating environment. I believe the passives are poised to follow
Customers are starting to value-engineer out capacitor content or switch to tantalum capacitors given the price squeeze in MLCCs, something that was not a historical focus for many of these components given that they were so cheap and generally only a small part of a typical bill-of-materials
Distributor inventory has risen to 6-7 months versus historical averages of 2-3 months, a severe buildup of inventory. The double-ordering and inventory buffering is beginning to correct and work in the opposite direction
Yageo's monthly revenue y/y trend peaked in August 2018 and has weakened sequentially every month since
I believe EPS is now poised to revert back to a more normalized level of NT$30 per share over the next few years as the MLCC shortage peaks out and declines begin in 2H19. At 8x normalized P/E, that is a NT$240 share price, or 30% down.
I do not hold a position with the issuer such as employment, directorship, or consultancy. I and/or others I advise do not hold a material investment in the issuer's securities.