subscribe: Daily Newsletter

 

Will Sony’s move make it even stronger?

0 comments

Sony has announced a plan to spin out its semiconductor business to a new entity, called Sony Semiconductor Solutions.
The transfer, scheduled for completion by April 2016, is both expected and surprising at the same time, writes Brian O-Rourke, senior principal analyst: consumer devices and MEMS sensors at IHS.
Over the past two years, Sony has spun out its audio and video business (February 2015), and its TV and PC businesses (February 2014). However, the semiconductor spin out is different. While the TV, PC and A/V businesses had been struggling to some extent with increased competition and a slow-growing world economy, the semiconductor business has generated a significant amount of Sony’s recent revenues and profitability. The semiconductor business is highlighted by image sensors, a historically successful and profitable business for Sony.
Sony is obviously moving in a new direction, in order to ease planning, shorten time to market and increase profitability. What will the ultimate effect of this spin out have on Sony’s image sensor business?
Sony has been wildly successful in the growing, but also highly competitive image sensor market. This has been true going back to the 1980s, when Sony CCDs powered devices such as camcorders and corporate security cameras. Sony was able to transition smoothly as the market moved to CMOS image sensors for camera phones and tablets.
Today, it dominates the CMOS image sensor business, with Q2 2015 revenues of slightly over $1-billion, good for over 42% of the worldwide market, nearly three times the revenue of its closest competitor, Samsung Electronics.
In the face of this market dominance of image sensors, the question arises for Sony: Why interfere with a good thing? The current Sony corporate structure seemed to be working well enough to ensure image sensor success. How will the new Sony Semiconductor Solutions benefit the image sensor business? The answer may be in simplifying a corporate structure so that Sony can react more rapidly.
If Sony has had a weakness in the image sensor market, it has been exploiting new markets and getting new sensors to market quickly. A recent example is the release of CMOS sensors with 1.0-micron pixels, the next step in CMOS phone sensor design. Both OmniVision and Samsung have already announced 1.0-micron models, while Sony has not. Also, Sony has been slow to new, emerging applications such as automotive cameras. It has just entered the automotive space in 2015, while competitors such as ON Semiconductor (through its acquisition of Aptina) have dominated the market for years.
Sony has stated that one of the purposes of its semiconductor spin out strategy is the “to more rapidly adapt to their respective changing market environments and generate sustained growth.” And the old Sony structure was somewhat difficult for the image sensor business. In addition to its size being a barrier to quick decisions, the old system included Sony Semiconductor Corporation for image sensor manufacturing and Sony LSI Design for design operations. Under the new structure, both units will be combined under Sony Semiconductor Solutions.
At first glance, it seems unlikely that Sony could improve its already dominant position in the image sensor market. However, the spin out to a new organisation that allows Sony to respond more quickly to market changes, and speed time-to-market for new designs, may have precisely that result.