The potential downsides can, however, be mitigated in part through increased design effort and managerial prudence.
Yet more mergers
A wave of mergers and acquisitions is sweeping through the semiconductor industry, with deals reaching in to the tens of billions of dollars now seemingly common-place.
2016 saw the announcement that Japanese group SoftBank would purchase ARM holdings for £24bn, preceded by the news that Avago would purchase Broadcom for $37bn. The momentum of high value buyouts continues with the recent deal struck by Analog Devices to purchase longstanding chip manufacturer Linear Technology for nearly $15bn.
These are just a few examples of a trend which is reshaping a global industry – an industry in which electronics engineers around the world rely to provide parts for their product designs. The conglomerations within the semiconductor industry could negatively impact new designs in terms of component pricing, diversity of parts, and industry innovation.
Rising prices, rising pressure
Continued mergers are likely to lead to an increase in component pricing. As the number of competitors in the market falls, the incentive for the remaining OEMs to be competitive over pricing is reduced, and the negotiating position of purchasing managers weakened.
This has clear ramifications for BOM costs, with downstream effects on profit margins or product pricing. A greater push for design optimisation to counteract this effect is likely – engineers should move away from the more integrated, more expensive component options where possible. Increased development cost and time is a natural consequence.
Mergers could also lead to a reduction in the diversity of available parts. With fewer IC manufacturers comes less choice. Finding the “right part for the job” is likely to become more difficult for designers, with performance compromises and additional components a necessary evil.
This is of particular concern to IoT device design, where device size and power consumption are at a premium. In addition, a degree of design homogenisation is to be expected, with product differentiation becoming more of a challenge.
Furthermore, such acquisitions could increase the rate of component obsolescence. As engineers watch familiar logos disappear from the top of datasheets, previously “safe-bet” parts are suddenly drawn into question. This effect is most likely to be seen in the short term following a merger, as superfluous components are dropped from catalogues.
Such concerns are brought into sharp focus in risk averse sectors such as medical, space and defence, and particularly in low and medium volume product designs. New and existing designs may require part substitutions and alterations, incurring the costs associated with part re-qualification. Redesign can be avoided by taking advantage of last-time-buy offers, although long-term storage of parts will incur costs of its own.
Practicalities over pricing and component management aside, there is perhaps a more profound effect to consider – a reduction in innovation. As competition is lost from the market, so is the drive for remaining companies to innovate. With such mergers there is also, to some degree, functional overlap between the two organisations and their employees.
Analog Devices’ buyout of Linear Technology, for example; despite largely expanding rather than repeating the current Linear Tech offering, there remains some repetition of function, for instance in power management and data conversion ICs. Should such an overlap translate into redundancies, the industry as whole will lose talent and design effort.
Keeping innovation alive
Corporate conglomerations within the semiconductor industry have the potential to negatively impact electronic design in the many sectors it serves, notably within the IoT. As the number of OEMs continues to fall, electronics engineers should be increasingly prudent with component selection for new designs. Shrewd team selection is also required to prevent these new design challenges from increasing product cost and development time.
Organisations operating within high-risk sectors can help manage the risk of obsolescence by planning ahead and analysing the mergers for indications of possible obsolescence rather than waiting for the EOL notices to be issued.
Product developers should aim to prevent a reduction in IC innovation from leading to a reduction in product innovation. Day to day, this requires engineers to step up to the challenge of keeping innovation alive at board level, rather than relying on the IC manufacturers to deliver it for them.
Gerry Hawkins is an electronic product development expert at PA Consulting Group.