Monday, April 22, 2013

Maybe Video Won't Crash the Mobile Internet


Bandwidth growth shows it is so difficult to accurately and consistently forecast the volume of change, even when the direction of change is clear enough. Everybody expects bandwidth consumption to keep growing. But we rarely get the magnitudes right.

Nor do we necessarily and normally hit the limits linear extrapolation suggests will be encountered, because actors behave rationally. Faced with higher prices for a product, they substitute other products, especially when suppliers encourage such substitution.

One might argue the potential explosion of video bandwidth will not happen in precisely the way observers and forecasters now predict, because users and suppliers will change their behavior. In other words, when there is no financial penalty for using bandwidth to watch video, people will watch. 

When there are incremental costs, behavior will change. And that is why, in the end, even mobile video will not crash the networks, though one could make a decent argument for that eventuality, extrapolating in a linear way from today's trends.

One might explain the Internet bubble demand forecasting failures. There were false signals being sent, in part because of fraudulent activity on the part of bandwidth sellers. But even when that is not the case, we tend to overestimate the degree of bandwidth demand growth.

At least one reason is that people and service providers have learned to act in ways that alter behavior. In other words, given service provider and end user self interest, mobile bandwidth growth has slowed because both suppliers and consumers benefit financially by doing so.

By 2017, almost 21 exabytes of mobile data traffic will be offloaded to the fixed network by means of Wi-Fi devices and femtocells each month, Cisco estimates. 4G Americas says Wi-Fi offload of mobile traffic is at 35 percent today in the United States and is estimated to be 68 percent by 2016.

Without Wi-Fi and femtocell offload, total mobile data traffic would grow at a compound annual growth rate of 74 percent between 2012 and 2017 (16-fold growth), instead of the projected 66 percent CAGR (13-fold growth), 4G Americas says.

Cisco notes that tthe global average for daily data consumption over Wi-Fi is four times that of cellular, averaging 55 MBytes per day for Wi-Fi, and 13 MBytes for cellular.


 Average Daily Wi-Fi and Mobile Data Consumption


Some think the same sort of trend ultimately will characterize mobile broadband bandwidth growth rates as well, In fact, there is little reason to doubt that future trend, given historical precedents.

In March 2011, for example, AT&T projected that data bandwidth growth would be on the order of eight to 10 times over then-current levels between the end of 2010 and the end of 2015.

That forecast appears to be based on an expectation that volumes would roughly double in 2011 and then increase by a further 65 percent in 2012.

Instead, AT&T seems to be seeing something like 40 percent annual growth. To be sure, 40 percent annual growth is significant. It means bandwidth consumption doubles about every two to three years.

Cisco estimates mobile broadband grew about 70 percent in 2012, and will grow at a compound annual growth rate of 66 percent from 2012 to 2017.

Some believe Wi-Fi offload will slow the rate of mobile broadband growth. On the other hand, even such offloading, at high rates of perhaps 80 percent, would slow the rate of growth by about 50 percent.


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...