Virulent Word of Mouse

October 21, 2011

US Broadband in Maps, Graphs, and some Bars

Filed under: Broadband,Internet economics and communications policy,Maps — Shane Greenstein @ 10:34 am

To be sure, most of us do not use government statistical reports as anything more than bedtime reading for inducing soporific reactions. It is cheaper than a sleeping pill.

But those expectations would be too harsh for the most recent broadband report from the FCC. It contains a great deal of data, and it is really quite informative. I would go even further. It is a useful vehicle for learning about the basic economics of broadband. For that purpose, however, it has one drawback: it is a wee bit too long, as in 88 pages.

This post will save you some time. Much of the key insights can be summarized in three pictures — a map, a graph and some bars. The post  will start with the map, then go to the graph, then end with the bars. (For those keeping score at home these pictures are taken from pages 62, 78 and 79 of the report.)


Let’s start with the map. It shows the distribution of broadband lines per capita at the census tract, where a line must exceed 765 kbps download and 200 kbps upload.

The map has one obvious visually defining trait: it is darkest in major cities. If a map of the concentration of US population were super-imposed on this map, then the pattern would be nearly a match. Every major US city contains the darkest areas.

In other words, the density of supply and use of broadband increases with the density of human settlement. Or, restating the policy issue, thin or inadequate supply of broadband largely arises outside of population centers, namely, in the rural areas of the United States. To be sure, millions of residences live in such areas, just not the majority of the US population.

Like it or not, that is the way it is. One picture summarizes the geographic feature of this policy issue in US broadband policy.


Now consider the next graph. It shows the use of broadband and market share for  different modes of broadband as a function of density. Blue is DSL, red is cable, green is fiber, purple is satellite, and orange is fixed wireless (i.e., a WISP).

(Aside: In this case, broadband is defined as 200 kbps in either direction, which is much more forgiving than the prior map. That is just a detail for a picture like this. I doubt the qualitative features of the bars would differ for another definition of broadband).

In many respects this graph summarizes the same economics as the map. However, there is more. It also shows something that was not obvious from the map.

First, it shows that the total use of broadband increases with density. That is most visible as we go from rural settings of low density (the far left) to suburban setting with moderate density (the middle).

But one other thing also comes through rather clearly: use of broadband decreases with density at higher levels (to the right of the middle). As density increases from suburban levels (the middle) to densely urban downtown levels (the far right), the use of broadband declines.

That happens, presumably, because more than density shapes demand, such as a buyer’s income. Income and density are not perfectly correlated. That is, not all the riches people live in the densest parts of the city, nor do all the poorest people live in rural settings.

Finally, look again at the far left. It has a feature not found on the far right. The wireless solutions are used in the less dense settings and almost nowhere else — presumably because the fixed provisions are not available.

Once again, one picture describes US broadband use in a nutshell, showing how an essential quality — technology choice — varies with density.


Those observations motivate the third and final picture, which is a bar chart. It looks at subscribers as a function of density and income.

(Aside again: In this case, once again, broadband is defined as 200 kbps in either direction. Once again, I doubt the qualitative features of the bars would differ for another definition of broadband).

This bar chart shows that income and density interact in a very interesting way. Both matter.

More precisely, density’s effect depends on income, presumably because the density lowers cost of supplying broadband. Income also matters, because it shapes the willingness to pay for broadband. Big drivers of supply and demand together in one picture, that is what this bar chart does.

So there you have US broadband in a nutshell, illustrated by a map, a graph, and some bars. Sometimes it is impressive how much information a few pictures can contain and convey.

What else did you notice?


  1. Thanks for the gloss on the report; I also found the figures about number of providers at various speeds quite interesting (p 9), far fewer than I would have expected. I realize this is out of scope for the report, but I would have loved to see some data about cost to consumer, and how those costs might compare to relative costs borne by consumers in other countries. Likewise how broadband penetration in U.S. compares to penetration in EU and elsewhere.

    Comment by Claire Stewart (@claireystew) — October 21, 2011 @ 1:35 pm | Reply

RSS feed for comments on this post.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s

Blog at

%d bloggers like this: