Qualcomm (NASDAQ:QCOM), the world's largest mobile chip maker, is currently mulling a split of its chip-making and patent-licensing businesses. Could Intel (NASDAQ:INTC), the world's largest maker of PC and server chips, step up and acquire Qualcomm's mobile chip-making unit to become a chip-making superpower?
Several Wall Street analysts believe that it's possible. Cowen and Co. analyst Timothy Arcuri told Reuters that it would be a "chip deal to end all chip deals." Ascendiant Capital Markets analyst Cody Acree stated that a merger would give Intel much-needed "diversity away from PCs." Let's weigh the pros and cons of a potential megamerger and discuss whether or not Qualcomm will split its businesses.
Why a merger might work
If Intel buys Qualcomm's chip-making business, it would instantly dominate the market for mobile application processors and wireless modems. Last year, Qualcomm controlled 51% of the mobile application processor market and 64% of the wireless modem market, according to Strategy Analytics.
ARM Holdings-licensed processor designs, like Qualcomm's, power 95% of smartphones and the majority of tablets worldwide. As for wireless modems, Intel controlled just 1% of the market last year.
To gain ground in mobile devices, Intel has subsidized OEMs with steep discounts on chips, co-marketing agreements, and financial assistance in redesigning logic boards for its chips. Those subsidies, known as "contra revenues," caused Intel's mobile operating loss to widen from $3.1 billion in 2013 to $4.2 billion last year. Buying Qualcomm's chip-making unit could eliminate the need for such subsidies.
Merging the two units could also cut the manufacturing costs for Qualcomm's chips, which are fabricated by third-party foundries like TSMC and Samsung. Since Intel has its own foundry operations, those partners would no longer be necessary after the merger.
Both Qualcomm and Intel are attempting to expand into the Internet of Things (IoT) market, which connects everyday objects -- like wearables, appliances, and vehicles -- to each other. Qualcomm and Intel both offer new chips for those products, but they use conflicting communication standards.
Qualcomm leads the Allseen Alliance, a consortium that backs its AllJoyn software framework for IoT devices. Intel backs the Open Interconnect Consortium (OIC), which supports the rival IoTivity framework. Merging Intel and Qualcomm's IoT operations would bring those chips under a single communication standard and potentially accelerate the growth of the IoT market.
Why a merger won't work
But investors should remember that Intel and Qualcomm both face major headwinds in their core markets. Back in March, Intel slashed its full-year sales guidance by nearly $1 billion, due to a slowdown in the PC market. In April, Qualcomm reduced its full-year sales guidance by over $1 billion after Samsung replaced Qualcomm's processors and wireless modems with its own components in the newest Galaxy devices.
Qualcomm's chip-making business is highly dependent on orders from the top few smartphone and tablet vendors. Qualcomm is also losing ground to smaller Asian rivals like MediaTek and Rockchip, which sell ARM-based mobile chips at lower prices. As a result, Qualcomm's chip-making revenue fell 22% annually last quarter and the unit's earnings before taxes plunged 74%.
The estimated value of Qualcomm's chip-making business is $30 billion to $40 billion. That's a hefty price tag for a business that's heavily dependent on two main customers, losing ground to cheaper rivals, and posting top- and bottom-line declines. It would also be Intel's largest acquisition ever, dwarfing its $16.7 billion acquisition of Altera.
Will a split even happen?
These pro and con arguments are meaningless if Qualcomm never splits up. The main problem is that Qualcomm's chip-making business is much less profitable than its patent-licensing one. The licensing business generated just a third of Qualcomm's revenue last quarter, but it raked in 85% of its segment profit.
By contrast, the chip-making business, which generated the remaining two-thirds of its revenue, only accounted for 15% of segment profit. As a result, Qualcomm supports R&D at its chip-making unit with patent-licensing profits.
Drexel Hamilton analyst Richard Whittington told Reuters that he couldn't see how Qualcomm could "possibly segregate the two" units and that short-term gains through a sale would be offset by "a permanent loss of market share" in mobile devices, especially after more of its wireless patents expire. Whittington suggested that Qualcomm should partner with Intel or smaller Asian companies instead of splitting its business in two.
I believe that the idea of Intel evolving into a chip-making superpower by absorbing Qualcomm's chip-making unit is wishful thinking. First, it's doubtful that a split will even occur, despite recent activist pressure from Jana Partners. Even if Intel were to buy the business after a split, it would expose itself to the fierce competition that is eating up Qualcomm's revenue, profit, and market share.