Finisar Could be Telegraphing 5G Fronthaul Gear

Avatar photoPublished on
Blog 1536426809 220

fibeReality is aware of some speculation that Finisar’s CEO, Michael Hurlston, was supposedly brought in to build up the 3D-sensing business in order to sell it to his former company, Broadcom. Yet, with our expectation of Gross Margins (GMs) in the business inevitably plunging, as we have referenced in the past, Broadcom would be even less inclined to become a player in this space than it was from the beginning, especially when it did not fit into its mandatory requirement of having GMs in the 60 percent range. While we also believe that it probably would not be the worst idea in the world for Finisar to make penetration into the traditional VCSEL bare die space, the possibility of greater inroads by II-VI in that business would make it a less attractive option. When it came to 5G wireless internationally, although Finisar’s former CEO, Jerry Rawls, did not hesitate to point out the benefits of the burgeoning, international growth in the future, we think a reasonable assumption was that the component vendor was really inclined at the time to take advantage of the need for additional metro type of bandwidth. Now we have a preliminary indication that Hurlston’s wireless experience (along with his industry contacts) could be potentially the most helpful in a new area of corporate focus for Finisar, which would be directed further down in the network – the access portion. At a bare minimum, his use of the term, “short…reach,” makes the firm’s strategy ambiguous. In particular, if Finisar were to become the lead player in helping Verizon set up a viable ecosystem for optical transceivers for 5G fronthaul (as well as backhaul requirements), especially at the chip level, it would not only set it apart from the Lumentum/Oclaro combination, but it would be a means to avoid being as dependent on the uncertain arrival of 400GbE. Of course, there are the difficult technical and large capital requirements at the higher speed, only to then inevitably become even more of a cutthroat situation than previous data rates without the LR differentiator.

Matthias Berger, formerly one of Nokia’s lead optical technologists (along with his quite talented technical team) was driving a fairly forward-leaning approach to cover fronthaul, but there was unwillingness by the system vendor as well as by a couple of component firms to put up the money to get to the desired price points. We surmise that was the reason for Berger joining Finisar within about the last month. (One should note that Hurlston had some experience with mobile backhaul apps in general using wireless mesh, and Finisar hired a senior architect with fronthaul/backhaul knowledge last summer.)

While Berger’s concept itself does not appear to be a close match with the NG-PON2 direction of Verizon, having him onboard at Finisar could turn out to be a coup. We remain convinced that the service provider is more than willing to go off-budget to ensure a viable ecosystem for the new access device.

As when it came to FiOS, Verizon’s preference is to principally deal with the smallest number of prominent vendors across the board to get the most flexibility on cost in building out its 5G network, including Ericsson/Calix for infrastructure as well as Corning for fiber and ancillary devices. In the same way, Finisar could be put in charge of the enabling componentry for NG-PON2, including consolidating (via acquisitions) some of the quite small chip vendors, which have been concentrating their efforts on this technology.

Finisar could also take the lead in ensuring the adequate development of the laser chips, which are critical in leading to the standard working, and at least to some extent, alleviate that burden on the startups in this space. If Finisar were to assume such a role with Verizon, it would keep its exposure to China down to a manageable level in whatever capacity the vendor chooses to support that country’s 5G infrastructure.

In the past, we have heard that Verizon Ventures is prepared to lay out a lot of cash to enable the manufacturing of two million transceivers for fronthaul deployments in 2021. Finisar may find it hard to resist going after that opportunity.

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]