metamorworks - stock.adobe.com
Kioxia and Western Digital, two of the top five NAND makers globally, appear to be inching toward a merger, which would shrink an already shrinking group of flash memory providers. But analysts are skeptical the merger will happen, and, if it does, that it will have a dramatic effect on the market.
Three financial lenders of Kioxia are refinancing a $13.5 billion loan to support a potential merger of the two flash companies, according to a Bloomberg report earlier this month. This comes after months of speculation that the two companies were in talks to combine, and as NAND prices have plummeted mainly due to oversupply from the COVID-19 pandemic. The terms of the deal would give Western Digital (WD) the controlling share, at 50.5% of the company; WD's HDD business won't be affected, according to Bloomberg.
The storage and memory markets are in a slump, which could be a stimulus for merging or continuing consolidation, according to Thomas Coughlin, president of Coughlin Associates.
"Everyone's valuations are down, therefore the cost of doing a [merger] is less," he said.
WD's market cap of $20.3 billion in 2020 dropped to $14.25 billion in 2023. The NAND producer with the largest market cap, Samsung, went from $500 billion in 2020 to $335 billion in 2023. Kioxia is not a publicly traded company. Western Digital and Kioxia declined to comment on this story.
Possible isn't definite
But the merger speculation isn't new -- even to this year. Reports began circulating in 2021, just a few years after Toshiba Memory Corp. spun off its memory and NAND products in 2018 and later renamed it Kioxia.
While a better financial deal for WD may lead to a higher likelihood of a merger, it isn't set in stone, according to Jim Handy, general director and semiconductor analyst at Objective Analysis.
"I'm not sure it really benefits WD," Handy said. "A merger would dilute the profitability of the NAND business for WD."
Kioxia and WD jointly design and manufacture the BiCS NAND, which is used in both companies' SSD products. Kioxia built the fabrication plants and both companies share the capital expense of equipping and running the facilities. A deal between the companies allows WD to take up to 49% or less of the factory output, Handy said. During periods of oversupply, which happens as part of the supply-demand cycle for flash memory, WD can choose to take less NAND, leaving Kioxia to deal with the surplus.
The fabrication plant is the most expensive part of the co-ownership, and currently Kioxia owns 51% of the plant and WD owns 49%, Handy said. For WD to take over, it wouldn't be a huge amount of money from WD, relatively speaking, he said.
Thomas CoughlinPresident, Coughlin Associates
Indeed, Coughlin said he believes WD controlling ownership would be in the company's best interest.
"What makes the merger more possible this time is that WD would be the majority owner," Coughlin said.
A merger wouldn't shake things up all that much, according to analysts. Samsung currently makes up 32% of the market share. Kioxia makes up 21% and WD makes up 16%, according to research from Objective Analysis. The combined company would be in line with Samsung, the long-standing market leader.
But a WD-Kioxia merger wouldn't result in the combined market share of the two companies, according to Coughlin. Customers will rearrange purchasing to spread market share with other vendors to reduce dependency, he said.
Handy said, "The theory is that if the industry consolidates, it will cause prices to normalize." He added that he doesn't agree with this theory.
Handy pointed to the DRAM market as an example. In the 1980s, there were more than 20 DRAM suppliers, and only six by 2012, which then consolidated down to three main suppliers. Customers still saw a collapse in price last year, Handy said. In a competitive environment, there will always be multiple factors to consider for price stabilization, aside from the number of vendors, he said.
Adam Armstrong is a TechTarget Editorial news writer covering file and block storage hardware, and private clouds. He previously worked at StorageReview.com.