Friday, September 16, 2016

Telecom Capex is a Problem: Spending as a Percentage of Revenue Has to Drop

Capital investments by telcos globally are growing slightly, and that might not be a good thing, as much as investment underpins creation of next-generation networks.


The problem is that service provider revenues are not keeping pace with investment levels. That is one reason why Facebook and Google efforts to create lower-cost access network platforms are important: ISP cost needs to decline. If all facilities-based service providers can use the new platforms, they win.


On a global basis, telecom service provider revenues grew about one percent in the second quarter of 2016, year over year.


That growth is the first for service providers since the third quarter of 2014.


Industry capex over the last 12 months was roughly $340 billion, flat versus the prior two years.


Revenues have been falling, with the result that service capex levels are historically high, at about 20 percent of revenues.


The trick is understanding how to balance capital investments between acquisitions and network investments. In many instances, revenue or profit might be more immediately affected, and to a greater degree, by acquiring assets, rather than investing in existing facilities.


That is one sense in which constantly-growing network capex is a “bad thing.” Money might more profitably be spent elsewhere.


The other issue is the cost of next-generation networks, in circumstances where all legacy revenue sources are declining. It is simple common sense that investments must match revenue expectations. If revenue is falling, or flat, so must capex fall or remain flat, over time.


As a practical matter, capex often is “lumpy” and “staircase” shaped, as investments are made to upgrade, then ratcheted back, and then boosted again to gain the next important increment of network capabilities.


And even if telecom has been a standards-driven and regulation-heavy industry for half a century, at least, a greater percentage of those standards might be created in new ways in the future, as a way of attacking costs.


To wit, competing carriers might begin to identify essential functions that do not confer competitive advantage, and collaborate with key rivals to lower costs in such areas.


Many carriers already have sold their mobile towers, or share tower costs, for example.


A few have outsourced operations tasks not seen as providing the ability to differentiate. In many markets, wholesale networks are the way retailers get use of those functions.


That sometimes is seen as the capital-efficient way to do things. Sometimes that is seen as the only way to boost competition when investment in facilities by other contenders is deemed extremely unlikely.


By definition, that approach fundamentally leads to commoditization of the access function, as all suppliers have the same speeds and quality of service.


So there are strategic reasons why various service providers might not want to open source or otherwise share business functions. Most importantly, every operator will try and gain some source of distinctiveness in their offers, if possible.

The trick is figuring out which essential functions offer the possibility of creating business value, and which do not; which functions are essential and which are not.

No comments:

Many Winners and Losers from Generative AI

Perhaps there is no contradiction between low historical total factor annual productivity gains and high expected generative artificial inte...