BLOG

Nokia’s Unchanged Strategy to Exit Optics

October, 2018

Even prior to closing the deal, Nokia Networks’ purchase of Alcatel-Lucent surrounded the former contemplating divesting the submarine optics division of the latter. The major obstacle with Nokia appeared to always be that the government in France would only accept a company in that country to purchase the assets. The goal of ridding itself of Alcatel Submarine Networks (ASN) first, has apparently remained in place. However, at first glance, from simply a financial perspective, with the boom created by the hyperscale data center operators, having stuck to this plan the whole time would not seem to necessarily make sense. In addition, the breadth of assets required to do new, very long-length undersea construction, would have made it even harder to unload. After reflection on the subject, we suspect that Nokia always understood that by eliminating the linchpin, the rest of its optical business would tend to collapse, and make it easier to discard. For example, fibeReality has discussed previously the close relationship between wet and dry long-haul equipment (which only got tighter with the move toward common standards). There has also probably been inadequate appreciation by many industry observers about the visceral amount of aversion at the top corporate level of being forced to play in the overall optical arena it abandoned before the big merger, and thus, it could not be fully grasped the degree to which purging this technology trumped bottom line results. So, before a potential buyout of the submarine product line by Ekinops was announced this month, the idea was to make the fiber optic effort in general, a secondary concern at best. We certainly addressed the inability of AlcaLu optical executives to acquire authority at the highest level of the corporation.

Consequently, despite owning the FiOS GPON market previously, Nokia showed insufficient interest in going down the path with Verizon on NG-PON2 (although there has undoubtedly been a long-term prejudice by the operator towards the Finnish entity, it may still be the second option after Ericsson/Calix). Yet, even with the emergence of 5G being anticipated for many years, prior to the AlcaLu acquisition, Nokia was obviously willing to sacrifice any expected advantages of having internal sources of optical access and metro equipment to support these next-gen wireless projects.

Nokia was also noticeably slow with R&D work to provide solutions best suited to the high-growth, DCI terrestrial space. Actually, the industry will never know what would have happened if Nokia had taken a hard stance against open line systems with the hyperscale operators, which almost assuredly would have been backed by many AlcaLu execs. Ciena would have likely been incentivized to do the same (Infinera would have supported the stance by default), and then the other system suppliers would almost certainly have gone along, so that at least some of the damage to margins across the supply chain could have been avoided.

Quite the contrary, fibeReality has reason to believe Nokia’s mentality may actually be for a push toward the opposite extreme in playing a part in changing the paradigm, using its DSP technology, to at least theoretically eliminate much of the need for a DCI box in metro applications. By employing the 400G ZR, it would allow for the data rate to go right into a mux/demux. Still, while we expect that there will be efforts at simplifying the DCI, our expectation right now is that it will still be necessary in many cases, and fitting it into a small, pluggable form factor will not inevitably be easy (including doing the necessary cooling), and so the current requirement for CWDM4, as a go-between, could be around for quite a long time.

Getting back to the wet side of the network, a reasonable conclusion is that Nokia, as the presumptive top supplier in the space, may have not have put up much of a fight in allowing the hyperscalers to drive prices down to a detrimental degree (at least in previous years, the major firms, supplying submarine projects would tend to be in the same ballpark on price). How else does one explain TE Connectivity divesting TE SubCom because of “a minimal contribution to profitability?” Obviously, the valuation of ASN has reached a point in which it allows even for a small system house manufacturer in France to consider making a purchase. (It should also be noted that before the closing of the merger, there was an aberration as AlcaLu apparently made a concerted effort to grab the maximum amount of market share with aggressively low pricing, with the notion that the new parent company would sell the subsea operation, and so, full recovery by the firm in not receiving a reasonable offer from such actions would have been difficult.)

Separately, it looks like the hyperscalers may have helped destroy another large, high-end ecosystem in the optics realm, except it involves mainly full systems this time. If so, are they now willing to also put up the capital essential for all of the ships, equipment, maintenance etc.? Once again, is not as if even Huawei Marine Networks (as opposed to Huawei Technologies on the terrestrial side) would likely be interested in providing such hardware and services on a subsidized basis in the same way that Amazon almost assuredly cannot bet the farm on just Chinese companies for all of the necessary optical component development, as well as volume production, in the quickest manner possible.

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]

 

Share Article

Add Comment

 

Back

 

Read Comments

  • thomas fay  
    October 29, 2018 at 10:56 AM Reply
    https://finance.yahoo.com/news/telia-carrier-infinera-demonstrate-first-070000411.html
    • thomas fay  
      October 30, 2018 at 15:35 PM Reply
      In other words, ICE5 is happening. Your comments, please.
      • Mark Lutkowitz  
        October 30, 2018 at 15:43 PM Reply
        It is a trial. Why is Infinera talking so much about the use of L-band on ICE4, even though that capability is not fully ready?
        • thomas fay  
          November 02, 2018 at 15:01 PM Reply
          https://www.infinera.com/wp-content/uploads/infinera-tp-demonstration-integrated-1.2-tbps-pics-transceivers-l-band.pdf they can do it and do it with PICS