II-VI: Disentangling Finisar at Both Ends

While there was initially hope on the part of at least some executives at Finisar with the new corporate leadership, it left a lot to be desired, which as it turned out, went well beyond messaging, when it came to running both its traditional and emerging optical businesses. Vis-à-vis the former, there has been a large amount of frustration by managers at the firm in splitting the company rather arbitrarily into separate units. With reference to the latter, the 3D Sensing (3DS) portion, fibeReality has discussed the fact that Finisar has been hemorrhaging critical VCSEL talent, and the principal reason for disenchantment, including by engineers still working for the manufacturer, is related to inexplicable noncompliance on a technological matter. In effect, we are fairly certain that Apple is quite alarmed by this state of affairs related to Finisar’s 3DS operations, as it had benefitted from its association with such technologists from the optical firm on design work of all kinds, well before the delivery of any VCSEL chips, and hence, likely urged II-VI to make this acquisition to fix the situation. Of far less importance, but it still may have been a factor, is Apple’s possible concern about its internal optical network, and that Finisar’s actions would have been detrimental to future development efforts on transceivers (although given the way 100G components were slammed down by the hyperscalers on price, we doubt that II-VI will be in any more of a hurry to move to 400G, despite rhetoric on the merger conference call of “even beyond a terabit”). Generally speaking, there would have been great uncertainty as to whether the leading data communications components vendor would have been able to get its act together by itself, at least in the short-to-intermediate terms. Yet, to be perfectly clear, we continue to anticipate that Apple’s plans for 3DS are tremendously more impressive than just the iPhone product line. Furthermore, in our opinion, the future success of the Sherman, TX fab remains vitally important to this customer.

Actually, the extent of the problematic overlap between the artificial, perforated lines created by Finisar, and the difficulties on the 3DS part of the business, is uncertain. The basic idea was to break up the company into pieces, and potentially sell them off.

However, in doing so, a lot of synergies get thrown away in that the company is not fully taking advantage of the strengths of its personnel. With Finisar having gone down this road for a while, it may be doubtful as to whether II-VI can undo all of the damage, even as the latter might be expected to change such a destructive strategy, and even if it decides to try to aggressively divest many of the acquired firm’s less attractive assets.

The initially important move for II-VI will be to totally reverse the momentum at the Sherman facility toward a process control structure, which is currently uncompetitive with Lumentum/WIN Semiconductors. Given that the latter combination is employing the standard configuration used by any other microwave/Gallium Arsenide IC foundries coming on line with VCSEL production, Finisar would be at disadvantage against them as well. We understand that the differences with Finisar’s manufacturing process order are minor, which makes resistance towards the necessary alteration seem even more incredulous.

In the meantime, VCSEL engineers there have been annoyed to say the least, as we suspect it has not helped the supplier to get its fab in Sherman up and running as quickly, and it surely avoids being fully aligned with the specifications being demanded by Apple of all of its vendors. If II-VI can totally get the plant in order, including possibly bringing back those valuable VCSEL engineers who have left Finisar (even if it means paying them a hefty premium), we project that it would be positioned in the long-term to be the dominant supplier of 3DS wafers to Apple.

This buyer would prefer to take advantage of a vertically integrated facility in one US location to maintain better control over its 3DS IP, to point to the factory, which it has already established ties from a political perspective, and then declare that it is getting the bulk of these components built in the States. In addition, with Lumentum’s propensity to do just about everything externally, including even the reliability work, its cost structure will inherently lead to a more expensive chip.

As always, fibeReality does not recommend any securities, and this writer does not invest in any companies being analyzed by us.

To follow us on our totally separate, quick update company blog, which is exclusively on fibeReality’s LinkedIn page, please click here.

[written by Mark Lutkowitz]