The NY Times had an editorial today on broadband and one of the faculty of Harvard Law School also wrote an op ed piece. In none of these polemics has anyone addressed any facts! Sometime and somewhere the facts should be included. So let me take another opportunity to relay them to those dreamers who think that this can be done quickly and painlessly.
1. First, the cost of fiber is as follows:
a. The cost of aerial construction is about $4500 per mile. That is fairly low and it assumes that the pole is pristine clean and there are no other costs and that aerial is all we need. That is also just the labor.
b. The cost for buried is almost $30,000 per mile, again just for labor.
c. The cost for make-ready, namely the cost to fix the pole so that the space one uses for simple aerial is $21,000 per mile. That is the cost to have the power and telcos move their stuff.
Now assume that we have say 20% buried, 50% make ready, and 80% aerial. The total labor costs per mile are $6000 plus $8400 plus $3600 or $18,000 per mile. If one were to borrow this money say from RUS with a 5.5% annual interest and a twelve year pay back one would pay $180 per month per mile for the fiber construction costs alone. That is about 1% of the CAPEX per month.
Now let us look at the cost of the fiber. The fiber is about $4000 per mile or $40 per month per mile.
The added electronics is about another $500 per subscriber or $5 per month per sub.
Now we have the drop costs of $200 per subscriber, the electronics of $500 per subscriber and the in home electronics for a 2 television set home of $1000 per subscriber. Thus the subscriber per home costs are $1700 per subscriber. However the per month costs are much higher since the lifetime of the electronics is six years. Namely it is 1.6% of the CAPEX per month or $27.20 per month per subscriber.
Now let us look at a simple middle class suburban neighborhood of homes with 200 feet of frontage, namely 200 feet from property line to property line. When one adds in streets and dead space that amounts to 300 feet or about 36 homes per mile. That is pretty dense for a lot of America. You see many are well below twenty but let us use 36. Then assume we have 50% penetration, a real wild optimistic guess, you will see why soon. That is 18 subs per mile.
The full fiber costs were $22,000 per mile or now $1225 per sub plus $1700 per sub for a drop and equipment for a total of $2975 per sub. This is an optimistic number.
Now the costs per month for the system alone is:
a. For the fiber and with 18 subs per mile we have $10 per sub per month for the construction, plus $2.25 per month per sub for the fiber or a total of $12.25 per month per sub for the stuff on poles.
b. The for the equipment we have $27.20 per sub per month.
Thus we have $39.45 per sub per month just for the CAPEX alone!
Now for the other costs.
2. The monthly operating costs fall into several categories.
a. The pole attachments are $2 per month per pole. There is about 120 feet between poles so each home has two poles and at 50% penetration this is $4.00 per month per sub for pole attachments.
b. Equipment maintenance is 5-10% per annum or say 0.5% of the CAPEX per month. There is approximately $3000 CAPEX per sub and this is reflected in a $15 per month per sub for maintenance costs.
c. The next cost is the Internet transit costs, the cost to peer with the Internet backbone. It is about $10-20 per Mbps per month. That means that if you were to stream at 1 Mbps 24 hours a day each and every day it would cost that much. But no one does that, they go on and off. So let's do a simple calculation. Let us assume that someone is on 30% of the time, and that when they are on they are transmitting or receiving 20% of that time and lest us assume they do so at 25 Mbps. That is a heavy user. This equals 1.5 Mbps as a load per sub. Assume we get a good deal at $10 per Mbps per month so each sub costs $15 per month.
Thus the total monthly cost of service is $4.00 plus $15.00 plus $15.00 or $34.00.
We can add them to yield, $73.45 per month per sub, before any operating costs! Now these numbers may vary depending on density of HH and penetration. Higher density and penetration will make some better but not all. A triple play will make things much better and an overbuild on an existing system, say a Verizon overbuild, may make them even better. But from whole cloth you really face an up hill battle.
3. Now add operating costs. They are as follows:
a. Billing is $2.00 per month per sub
b. Network Management is $1.00 per month per sub
c. Customer Care is $2.00 per month per sub
d. Overhead is $2.00 per month per sub.
That is $7.00 per month per sub for OPEX!
The total is $80.45 per sub per month! That is a lot. How can cable companies do it, the secret is video and voice. The CAPEX is marginal and the OPEX is shared. The COS is also somewhat shared.
So how do these industry opinion leaders deal with these facts. Well frankly they don't, for they appear to be clueless. We did these number when we obtained our RUS funding five years ago and it is a risky business. Two competitors is about all the market can stand. That is what we have today.
As to Amsterdam, it is totally different. It is dense in terms of housing, as is most of France, they lived in clustered areas as compared to the US. The US is different and thus more costly.
The US is geographically and demographically different than many countries, having lived from time to time in 23 of them. Thus when we look at the costs of broadband in isolation it is quite high. What one wonders is the role of Government in this area. Economics drives the deployment, not wishful thinking and politics, at least it should. Not one of these opinion writes seems to have the slightest bit of experience in this field yet they make statements for which they have no basis in fact.
At some time perhaps these opinion leaders may have to face the facts, but we suspect the facts are a bit above their arguments. These opinion makers want the Government to fund, build and operate the backbone! This was abandoned by almost all nations decades ago, yet they want to recreate it here!
Imagine if the world was run by engineers instead of lawyers and economists, facts would matter. Oh by the way, China is!