Only 8% Cost Difference between PON and Home Run Fiber Architecture?
April 28th, 2007 by Gordon Cook
Martin Geddes pointed out what he called an Interesting data point on PON vs Homerun
Herman Wagter who is carrying out the fiber build in Amsterdam responded:
To make a fine but relevant distinction: the original text claims an 8 % lower investment for PON compared to point-to-point. (Our estimates were of the same level of magnitude). If this will translate to a lower level of costs is almost entirely dependent on the depreciation time of the investment, and a quite risky bet.
To put it more bluntly : Who can be sure that PON has a lifecycle of 30 years, matching the growing demands of users? What is the risk that you have to switch to Point-to-point prematurely (say after 15-20 years) and throw away the remaining asset value? Or the risk that you have to change the electronics/lasers to a much more expensive setup (wavelengths) to match the growth? Do you want to pay an insurance premium of 8 % to mitigate that risk?
Our answer was: yes, we’ll take the insurance. The option value of Point-2-point is much higher than of PON.
(Note: one can use PON electronics in a point-to-point passive architecture (by using splitters in the CO), but not the other way around. So, if by some unforeseen technological development, PON becomes the dominant architecture, one can use the equipment anyway without a penalty. )
Gordon,
Herman raises an excellent point, but that was kind of my point too. I couldn’t understand why - with such a limited investment differential - incumbents were favouring PON. The answer seems to be that it’s an insurance against unbundling, certainly not an insurance against the lifecycle of PON.
I still think it’s a very short term view - not to mention that unbundling DSL has been profitable to most incumbents - and that should fiber services really take on, incumbents who are deploying PON may come to regret it.
Anywway, thanks for the link, I’ve added your excellent blog to the list on my own.