Broadband Expert Andrew Odlyzko Warns Telecom Investors That Industry Has Its Math Wrong, Again

Expert Opinion November 4th, 2014


November 4, 2014 – Nearly a year-and-a-half ago, Verizon Communications CEO Lowell McAdam claimed in an editorial in The New York Times, “The United States built its lead because companies invested nearly $1.2 trillion, over 17 years, to deploy next-generation broadband networks.”

That number is meaningless: over a period of 17 years, much of the infrastructure would have to be replaced at least once. Internet progressed from dialup to digital subscriber line (DSL) to cable and fiber today. Cell phones went through at least two generations of change in this time. McAdam may be quoting meaningless numbers because meaningful numbers don’t make Verizon look too good.

You can find meaningful numbers for investment and profitability in the latest paper by Professor Andrew Odlyzkoandrew-o, a mathematics professor at the University of Minnesota who has a history of getting internet statistics right when the rest of the world gets the numbers wrong. In 1998, when telecommunications industry boosters were claiming that traffic was doubling every 100 days, Odlyzko said it was growing by 75 percent to 150 percent each year. The difference between 1000 percent each year and 100 percent each year turned out to be the difference between telecom boom and telecom bust. He gets his numbers from authoritative sources such as the OECD and CTIA.

Today, when the cellular telephone industry claims that it is burning cash, Odlyzko says it’s not true. When the cellular telephone industry says its business is driven by demand for high bandwidth video, Odlyzko says that industry growth is driven by demand for communications, and that the interesting trend is that texts are replacing voice calls.

In his new paper, Will smart pricing finally take off? (available here), Odlyzko prints the following table:

Table 2: Voice to text substitution (US)

year voice minutes billions texts billions
2005 1,495 81
2006 1,798 159
2007 2,119 363
2008 2,203 1,005
2009 2,275 1,563
2010 2,241 2,052
2011 2,296 2,304
2012 2,300 2,190


The cell phone companies are pleased because texts are where the profits are, as explained by table 1, price per megabyte (the table shows approximate numbers, but those approximate numbers show where the profits are):

SMS $1,000.00
cellular voice 1.00
wireline voice 0.10
residential Internet 0.01
backbone Internet 0.0001


The reality is that cellular networks don’t need to invest in platinum-plated networks. Their customers merely require a network that can deliver text messages. Cell phone network companies (if you can believe their SEC filings) are incredibly profitable, and are spending relatively little on infrastructure:

year revenues in $ billions capex in $ billions capex/revenues
2004 102.1 27.9 27.3%
2005 113.5 25.2 22.2
2006 125.5 24.4 19.4
2007 138.9 21.1 15.2
2008 148.1 20.2 13.6
2009 152.6 20.4 13.3
2010 159.9 24.9 15.6
2011 169.8 25.3 14.9
2012 185.0 30.1 16.3


Odlyzko estimates that it would cost only $240 billion to replace the equipment in every cell phone network in the U.S. He reasons:

In wireless, industry statistics show that cumulative capital investment, from the start of service three decades ago, came to $365 billion by the end of 2012. Much of that investment has of course been written off, as old equipment gets replaced. So to replace everything (and it is far easier to replace telecom installations, even cell towers, than it is to replace electric power plants), would probably not cost more than half of the cumulative total, or about $180 billion. But just to be safe, let’s assume it would take $240 billion.

He thinks it is shocking that Verizon had to pay $130 billion to buy out Vodafone’s 45% share of Verizon Wireless. It makes sense only because “modern telecom is less about high capital investments and far more a game of territorial control, strategic alliances, services, and marketing, than of building a fixed infrastructure.” Verizon must be assuming that Vodafone’s high profit margins will continue but that is only possible if the cellular telephone market in the U.S. has already succumbed to the market failure known as monopoly.

This blog post does not describe the focus of the paper, which argues against an assumption by many industry analysts that telecommunications companies should charge based on usage instead of charging a flat fee. The numbers, which are a mere excerpt, are useful because once again, Odlyzko is warning the industry that it has its numbers wrong, and the last time Odlyzko had the numbers right and the telecommunications industry got its numbers wrong, telecommunications investors lost $1 trillion.

Alexander Goldman is a recent graduate of Brooklyn Law School, and recently passed the New York Bar Examination. He worked at ISP-Planet and ISPCON, was Chief Analyst for CTI’s American Recovery and Reinvestment Act grants, and had internships at the Federal Communications Commission and the Internet Division of the NY State Attorney General. At Brooklyn Law School, he was a Trade Secrets Fellow and won CALI awards in Contracts and Antitrust. accepts commentary from informed observers of the broadband scene. Please send pieces to The views reflected in Expert Opinion pieces do not necessarily reflect the views of and Broadband Census LLC.

You may also like:

Tagged with: , , , , , ,

One Response to “Broadband Expert Andrew Odlyzko Warns Telecom Investors That Industry Has Its Math Wrong, Again”

  1. Doc Searls Weblog · Huge Black Friday Tab Sale Says:

    […]… […]

Leave a Reply