Following the announcement from chipmaker Marvell (NASDAQ:MRVL) that it would be winding down its mobile platform business and slashing its headcount by 17%, analyst Christopher Rolland from FBR Capital Markets (via Barron's) noted that his team was "surprised by the lack of a buyer" for the mobile division.
I'd like to give three reasons I believe Marvell was unable to find a buyer for its ailing mobile platform business.
No. 1: It simply wasn't competitive
When trying to get a sense of how competitive a semiconductor company is, I like to look at its gross profit margins relative to its peers. Generally speaking, companies that ship leadership products and enjoy robust economies typically have higher margins, while companies that aren't as competitive and lack scale tend to have lower margins.
Marvell disclosed in the press release announcing its mobile wind-down that its gross profit margin in the mobile business during the first half of its fiscal year 2016 was just under 11%. This, to me, signals that the company's product portfolio was in such poor shape that in order to move product, it had to sell its platforms at prices barely above cost.
From this alone, it's not hard to see why Marvell wasn't able to sell the business. However, there's more.
No. 2: A potential buyer would need to lose money before making it
Any potential buyer of Marvell's mobile business would be taking on a business that is losing a lot of money. In its press release, Marvell estimates that it will be able to save somewhere between $170 million and $220 million in annual operating expenses as a result of the wind-down of this business -- giving us an idea of its annual mobile-platform operating expense run-rate.
Also note that the business generated, per Marvell, around $122 million in revenue and $13 million in gross profit. A quick-and-dirty estimate of Marvell's annualized losses in this segment, assuming similar revenue/gross profit in the second half of the year and the mid-point of the given operating expense range, comes out to a whopping $169 million per year.
In other words, a potential buyer for Marvell's mobile platform business would need to incur substantial losses associated with the asset for quite a while. Such a hypothetical buyer would need to really want to be in the mobile game, which brings me to my last point.
No. 3: Who would actually want to buy it?
The trend in the mobile-chip business, particularly over the last few years, has been that companies are choosing to exit the market, not enter it. This means that Marvell's mobile business probably didn't have much of a chance of going to a buyer that wanted to quickly enter the mobile-chip market.
Marvell's mobile platform unit also didn't have much of a chance of being acquired by a larger, more successful competitor in mobile chips. Indeed, from a technology standpoint I don't see what Marvell could offer, say, MediaTek or Qualcomm and if either of those two companies wanted Marvell's talent, they can hire some of them without needing to "buy" the entire business.
In a nutshell, Marvell probably had a hard time finding a buyer because there are very few (if any) companies at this point that would even be interested in buying any mobile platform businesses.