Index Home About Blog
From: "Fred R. Goldstein" <>
Subject: Re: Costs of Expanding Outside Telephone Plant
Date: 30 Apr 90 18:52:03 GMT
Organization: Digital Equipment Corp., Littleton MA USA

In article <>, (TELECOM 
Moderator) writes...

[with regard to why it costs too much for rural telcos to provide 
1-party service]

>...The wire
>lasts a long, long time. And once it is paid for, it keeps right on
>turning an almost 100 percent profit for its owners, allowing for what
>repair is required from time to time.

>I think the same consideration has to be given to party line service.
>The conversion would cost an arm and a leg, but if they plan it right,
>and install enough new cable to account for anticipated needs -- not
>for today -- but the next half century or so -- it will pay off.

Actually, it's not as easy as Patrick makes it sound.  The simple
arithmetic "payback time" is mathematically wrong, as it fails to take
into account interest rates.

A telco makes something like 12% return on investments.  That's set by
the state and FCC.  (Usually it's a bit higher.)  Rates are set in
order to generate that rate of return on the invested Rate Base, which
is the sum total of all nondepreciated capital investment, after
paying off all expenses (which includes depreciation).

How long would it take to pay off a home mortgage if you paid the bank
the equivalent payment of 5% interest, but it accumulated "negative
amortization" based on an interest rate of 10%?  It doesn't work.
We're asking telcos to do something that even a Texas S&L wouldn't
have tried!  :-)

If a telco can only recover 5% of the cost of something in a year,
then for every $1000 that goes into the rate base, $120 of Revenue
Requirement is generated, but that is matched by only $50 of income.
Thus the telco actually loses $70.  If the depreciation is, say,
20-year straight line (which is VERY slow, but some states do this
sort of thing to hold down the revenue requirement and thus local
rates), then $50/year is an expense, and that cancels all of the
income.  Thus there is exactly 0% return on investment for 20 years,
and $120 of Revenue Requirement that _will_ get made up elsewhere.

Rural telcos get far more than urban ones in their "separations" (now
collected via access charges for calls terminating from LD carries).
Sometimes AT&T pays the local telco twice as much as it charges the
caller!  This goes to subsidize the very expensive local plant.  Such
subsidies are responsible for the drastic reduction in party line
service that we've seen in the past two decades.

As a matter of public policy, the FCC has chosen to allow toll to
subsidize rural telephony in that way.  However, cost-based pricing is
gradually whittling down the subsidies.  We city slickers may not be
paying quite as much for Farmer Jones' line in the future.  Some
subsidies, however, will persist, as there is an explicit
toll-financed fund for that purpose as well as the "hidden" subsidies
in the toll rates.  

Fred R. Goldstein  or  
voice:  +1 508 486 7388 
opinions are mine alone; sharing requires permission

[Moderator's Note: Fred, the problem I have with your figures is that
if the circumstances were exactly as you describe them, how could
*any* major expenditure at a telco be justified?  What was the
justification for the millions spent converting old offices to ESS?
And what about ISDN?  The new technology is taking BIG $$ to install
and maintain. When is the payback? In some cases, years away. I
realize there is not a direct correlation between conversion from
party line to one party service and some of the other new-fangled
hi-tech stuff, but still -- why does any modernization go on at telcos
if the scenario is as grim as you paint it?  PT]

From: "Fred R. Goldstein" <>
Subject: Re: Costs of Expanding Outside Telephone Plant
Date: 3 May 90 17:35:08 GMT
Organization: Digital Equipment Corp., Littleton MA USA

In article <>, Donald E. Kimberlin, TNA, Safety
Harbor, FL (via MCIMail 413-3373) writes...

>Mssrs. Goldenstein and Wolfe, each in their own way, seem to pro-
>liferate a misconception oeprating telephone companies want us to
>hold: That the capital cost of building new transmission facilities
>with today's technology is as heavy a burden as it has always been.
>        Not true!  In fact, the base capital cost of providing added
>transmission facilities has plummeted in recent years, often to
>factors one-tenth or less of what they were even a decade ago ... and
>even in unadjusted-for-inflation figures.

Mister Kimburling (sic) seems to have a great faith in the ability of
technology to overcome economics.  Sometimes it's true.  Sometimes
it's not.

This thread began, if you recall, with a comment about rural
multi-party telephones.  Not city or suburb, but serious boondock
country where the local telco doesn't even ask you if you want
single-party lines.

In such cases, the latest and greatest telco transmission technology
can help a little, but not a lot.  This is an especially serious
problem in America, due to the rural land pattern.  In Europe, for
instance, most agricultural areas are arranged village/farm, where
everybody lives in the village and farmers commute (maybe less than a
mile) to their fields.  In America, small farmers tend to live on
their land, amongst the pigs and chickens.

In the homesteaded areas of the country (i.e., the Great Plains), land
tenure was created in units of 40 and 160 acres, so the houses tend to
be evenly spaced every half-mile or so.  Only the center of each
Township (usually a 6x6 mile square) has a village.  In northeastern
areas, rural areas are marked by natural boundaries (mountains,
rivers, etc.) and houses are scattered willy-nilly along winding

It's these customers who are hardest to serve.  When you are within a
road-mile of only three or ten neighbors, who is there to mux your
line with?  Muxes require a "star" configuration at the
pedestal-point.  Sure, that works in villages, and it allows the
telcos to substantially reduce the distances they have to go.  But the
"last mile" is a lonely, costly one, with maybe a 25-pair aerial cable
serving a few road miles.

I did run into one interesting technological advance aimed at such
markets.  A little start-up developed a fiber optic mux system with a
very low cost, weather-safe (mount atop the pole) terminal, designed
to work in a tree-topology exchange plant.  For various reasons it
isn't suitable for the US market and I don't know how they're doing,
but it's not like nobody's thinking about the problem.  It's just that
the really spread-out customer base in rural areas isn't easy to serve
with any available technology.

(Yes, there are a few technologies that have been suggested but for
various reasons, like the FCC, not approved.  Radio local loops (fixed
cellular is a fancy version) are sometimes useful.  Cable TV can carry
telephone channels too, if you solve the privacy issues (i.e.,
digitizing and encrypting).  But that still requires physical copper,
not cheap to pull.)  

Fred R. Goldstein
                    voice:  +1 508 486 7388 
disclaimer:  opinions are mine alone, sharing requires permission

From: "Fred R. Goldstein" <>
Subject: Re: The Mis-Named FCC-Mandated Charge
Date: 21 May 90 16:33:33 GMT
Organization: Digital Equipment Corp., Littleton MA USA

In article <>, (Will
Martin) writes...

>I note people on the list stating that they pay $3.00 even. First off,
>if this is an FCC-mandated surcharge, and thus national in scope, why
>is it higher here than elsewhere? What is the justification for it
>varying from region to region? Should it not be identical nationwide?

There's an FCC-imposed _cap_ on it, but the telco can charge less than
the cap for residential subscribers.  The FCC must approve the rate.
It tends to be higher in rural states, where the cost of local service
delivery is higher.

>Secondly, this charge has risen over its life. I seem to recall it
>started out at $1.50, though that may have been $2.00. I can
>understand the motivation behind the charge, to replace the revenue
>lost from kickbacks from LD service (I may not *like* it, but I can
>*understand* it... Grrrr... :-), but what possible excuse can there be
>for it having *risen*? The kickback-revenue was lost when the breakup
>and deregulation occurred, and the charge was instituted then. Over
>the ensuing years, it should have decreased, so as to be phased out,
>not increased. Who paid off who to get *this* gravy train?

The kickback from LD _calls_ has been reduced and continues to be
reduced, but the separations formulae which put much of the cost of
local telephone service into the interstate jurisdiction haven't been
reduced accordingly.  Thus the fixed charges (which replace part of
the cents per minute of LD calls) go up in order to make up the
difference.  It used to be that heavy callers subsidized light
callers.  That's less true now, though still true to some extent.

The gravy train leads to places like Wyoming and North Dakota, where
the average cost to deliver local telephone service is much higher
than average.  (Beehive Tel. in Utah invests $7000 per subscriber
line.  See Art Brothers' column in Telephone Engineer & Management.
The national average is more like $1000.)

In order to keep this flowing, separations cauases about 30% of the
non-traffic-sensitive (NTS) cost (not _price_) of local service to be
put under federal jurisdiction.  That's for the line, not for calls
(which would be traffic-sensitive).  Given an average NTS subscriber
line _cost_ of about $20/month, and a 30%ish federal share, the
$6/month cap is reasonable.  It's lower here in Mass. since the NTS
cost is only about $13/month, per NETel filings.

>Is the income from this surcharge treated differently, for accounting
>purposes, than the income from the "real" charges for telephone
>service by the BOCs/telcos? Or does it all just get dumped into the
>same pot?  (I have this image of the cellar of the new SW Bell
>building here looking like Scrooge McDuck's money vault... :-)

It's all "real" income, but this is FCC-regulated rather than
state-regulated.  The pots are handled differently; the federal pot is
skimmed for a special fund to subsidize high-cost rural telcos.
Otherwise, rates in the boonies would be a LOT higher than they are.
City rates would be a little lower.

>Lastly, is there any official plan for this charge to *ever* go away?

Of course not, since it's as much a part of your bill as the
state-regulated part!  The only way it would go away is if Congress
modified the Communications Act, and there's no good reason to do that
here.  The telcos are guaranteed a fair rate of return, and if they
didn't get it one way, they'd get it another way.  

Fred R. Goldstein   or
voice:  +1 508 486 7388 
opinions are mine alone; sharing requires permission

Subject: Costs of Expanding Outside Telephone Plant
Date: 28 Apr 90 15:32:32 EST (Sat)
From: Larry Lippman <kitty!>

In article <> (Irving Wolfe) writes:

   (quoting Peter da Silva)

> >What I don't understand...
> >Why would anyone be using a party line service in 1990?
> Because there are still garbage phone companies throughout the
> country, like Telephone Utilities here, who won't string new lines as
> an area grows unless they absolutely have to; even then, it takes
> forever.  Except for business lines at business rates, they are
> apparently free to say, "You can have a party line in a couple of
> weeks.  If you want a private line, we can put your name on a list to
> get one when someone relinquishes one."  When there's enough backlog
> to generate an instant payback on the new lines, they'll finally send
> a crew out to do something.

	I don't believe there is an operating telephone company
anywhere that does not share your desire to replace party lines.
Party lines require more complex CO apparatus, are a CO maintenance
headache, and are an outside plant maintenance headache since
generally when one subscriber goes down, all subscribers go down.
Many 4-party circuits still require ONI for toll call identification.

	What you totally fail to comprehend, however, is the
significant costs associated with extending outside cable plant.  Let
me give you an example of what it would cost to extend a 100-pair 24
AWG cable on an aerial route a total of JUST 2 MILES.  I am assuming
H88 loading because you probably wouldn't have party lines if you are
less than 20 kft from the CO, anyhow.

10,560 ft 	24 AWG 100-pr "figure-8" self-supporting
		cable					$  17,000

50 sets		Pole hardware				$   1,000

20 sets		Ready access boots & hardware		$   1,500

8 sets		Splice cases with splices		$   1,500

2 sets		H88 loading coils with cases		$   2,000
		Materials subtotal			$  23,000

200 man-hr	Installation of pole hardware and
		running of unterminated cable		$   4,000	

80 man-hr	Install 8 splice cases and 2 loading
		coil cases, and make splices		$   1,600

80 man-hr	Install 20 ready access boots		$   1,600

20 man-hr	Test all 100 pairs from CO to end of
		circuit, and create outside plant
		records					$     400

75 man-hr	CO work to rewire 50 subscribers and	
		assign to new line equipment		$   1,500

150 man-hr	Outside plant work to reterminate
		drop wires, cut-in pairs at ready access
		boots and rewire or replace 50 subscriber
		stations				$   3,000
		Labor subtotal				$  12,100

		Grand total				$  35,100 

		Investment for 50 subscribers, per line	$     702

NOTE:	The above labor rate is $ 20.00 per hour with some burden
	thrown in.  This may be reasonable for a smaller independent
	operating telephone company, but it is absurdly low for say
	a BOC or GTE.  In general, my estimate should be on the
	*low* side of actual cost.

	The above scenario assumes a 50% future growth, which is a
reasonable reserve for rural cable plant installation.  Revenue to
offset outside plant investment is assumed from 50 subscribers.  The
investment cost, based upon 50 subscribers is $ 702.00 per subscriber.

	How quickly do you think a small independent operating
telephone company can recoup the $ 702.00 per subscriber?  If the
party line service were forcibly withdrawn and monthly rates raised by
$ 10.00 per subscriber (which would create a minor revolution in most
communities!), it would still require SIX YEARS to recover the cost of

	Actually, it is neither reasonable to assume that all 50
subscribers can be made to subsidize the cost of investment, nor is it
reasonable to assume that rates can be increased by $ 10.00 per month.
Some subscribers will *insist* upon retaining the party line at party
line rates, so no new revenue will come from them.  The result is that
the return on investment period is closer to TEN YEARS.

	Even if the telephone company elected to make the outside
plant upgrade, it has to *borrow* the money from somewhere, usually
from the REA or through issue of bonds.  So now we have an interest
expense that I have not even considered.

	Many people have the false impression that an independent
operating telephone company is making oodles of money and that failure
to provide equipment and plant upgrades are the result of *greed*.
Wrong.  Times have changed.  Materials are expensive.  Labor and
burden is expensive.  Look at my example above, and tell me where the
money is going to come from?  Santa Claus?

> Does anyone know anything about starting a local telephone cooperative to
> wipe these bums out of business, or is that a pipe dream?

	You obviously feel that you know more than the "bums" running
your local independent operating telephone company.  Why don't you
simply tell them your "secret" of circumventing the economic realities
that I have outlined above?  With the benefit of your sage advice, I'm
certain they'll begin the job forthwith.

Larry Lippman @ Recognition Research Corp.  "Have you hugged your cat today?"
UUCP:    {boulder|decvax|rutgers|watmath}!!kitty!larry
TEL: 716/688-1231 || FAX: 716/741-9635      {utzoo|uunet}!/     \aerion!larry

Subject: More Comments on Costs of Expanding Outside Telephone Plant
Date: 8 May 90 13:58:56 EDT (Tue)
From: Larry Lippman <kitty!>

	In one fell swoop, I am going to address issues raised in
various articles which responded to my posting about the costs of
expanding outside telephone plant.

In article <> (Roy Smith)

> 	And then gives a cost breakdown totaling $35k, half of which
> is the cost of the copper-conductor cable itself.  What I'm wondering
> is if there is some cheaper way of doing it.  How much would it cost,
> for example, to run a single pair (possibly stealing an existing
> party-line pair) and run T2 over it, giving you 96 voice circuits (if
> I'm not mistaken)?  Obviously you need a mux at both ends (the
> SLC-96's discussed on this list a few months ago?) and power at the
> remote end to make it work, but it sounds like it might be a lot
> cheaper than the route Larry described.

	First of all, from the tone of the original poster's article,
I made the assumption that he was referring to an independent
operating telephone company in a largely rural area.  If we are going
to refer to a largely "suburban" rather than "rural" setting, then
much of what I have stated will not be applicable.

	There are *many* options available to effect "pair gain" in
outside telephone plant.  Unfortunately, for a largely rural CO
serving area with a generally uniform distribution of subscribers
beyond say, a 3 mile radius of the CO, *none* of the many pair gain
alternatives may be as practicable as simply expanding the cable
plant.  Let's look at some of these pair gain options and see why:

1.	One-channel subscriber line carrier, like Continental AML,
	SSL or equivalent, is an analog FDM system designed to
	piggyback one carrier subscriber on top of one physical
	subscriber.  This type of carrier cannot be used on loaded
	cable because of transmission loss at the FDM frequencies in
	the range of 28 kHz to 80 kHz.  Besides, this stuff is real
	crap.  We can rule out this option because of distance
	limitations, if no other reason.

2.	Four or six channel subscriber line carrier, like Continental
	CM-4, CM-6 or equivalent, is an analog FDM system designed to
	provide four to six subscriber line circuits on one pair of
	wires dedicated to the carrier circuit.  This type of carrier
	cannot be used on loaded cable because of transmission loss at
	the FDM frequencies in the range of 28 kHz to 150 kHz.  While
	there are repeaters available for this type of carrier, a system
	installation is generally limited to an overall loop resistance
	of 2,400 ohms or around 140 dB total loss at the highest carrier
	frequency.  Such loop restrictions quickly eliminate use in many
	rural environments.  This stuff is only marginally better than
	than AML in (1) above.  We can also rule out this option,
	primarily because of distance limitations.

3.	Analog space-division technology like the WECO 1A Line Concentrator
	is dead.

4.	Okay, state-of-the-art time.  Let's talk about digital subscriber
	line multiplex, line the WECO SLC-96 or its non-WECO equivalents
	(like apparatus made by Digital Telephone Systems).  This type
	of apparatus uses 2 to 6 T1 pairs, depending upon configuration
	and whether automatic span protection is employed.  We can obtain
	up to 96 subscriber lines from no more than 6 T1 lines using this
	approach.  Assume for the moment that I am referring to digital
	multiplex which is standalone in the CO; i.e., there is a physical
	loop terminal for each line in the CO multiplex apparatus which
	directly connects to the MDF for each and every working line.

	Sound pretty good so far?  Well, first question is: does it
	really help to provide a quantity of new subscriber lines at
	*one* specific location remote from the CO, bearing in mind
	that we have a rural situation with a generally uniform
	distribution of subscribers?  When one considers the additional
	cable necessary to route subscriber lines to the field location
	of this apparatus, it may no longer be so attractive.

	Second problem: we are now going T1 where T1 has not gone before,
	so what will we need?  Regenerators, spaced every 6,000 feet.
	Now we have a lot of cable work (splices, removing loading
	coils, multiples and bridge taps, etc.), cost of regenerators,
	and the cost of the multiplex apparatus itself.  This is no
	longer sounding so great.

	Third problem: if we are really in the boonies, we may not have
	regular alpeth cable, but we may instead have one or more strands
	of "REA cable".  REA cable typically consists of 12-pairs of
	19 AWG which are loosely twisted together about a suspension
	strand with individual pair color coding, but with no outer
	sheath.  Ain't no way to *reliably* run T1 through REA cable. 

	Fourth problem: digital subscriber line multiplex is not that
	inexpensive.  Assuming that the multiplex is of the variety which
	creates individual physical pair terminations in the CO (i.e.,
	not for direct T1 interface to a digital ESS switch), installed
	costs can range between $ 400.00 and $ 600.00 per line.  Actual
	costs depend upon the particular type of apparatus, the capacity
	of the apparatus, the number of initial working lines, remote
	site power options, etc.  Adding costs of cable preparatory
	work and installation of T1 regenerators may place the cost of
	this apparatus on par with expanding physical cable plant.

	Fifth problem: resistance to "new-fangled" technology from
	independent operating telephone companies still operating
	"old-fangled" CO's with SxS, XY or other electromechanical
	technology.  This may seem hard to believe to some Telecom
	readers, but there are independent operating telephone companies
	today who do not even own an oscilloscope, with their most
	sophisticated piece of test apparatus being an old TTS-4.
	How can such a company survive?  Simple, AT&T provides intertoll
	and DSA trunks using T1 or N-carrier and terminates them on their
	*own* apparatus frame in the independent telephone company's CO.
	So, AT&T does all the "electronic" troubleshooting in their own
	apparatus.  All of the independent telephone company's own
	apparatus may not have a single digital IC.

5.	The other type of digital subscriber line multiplex employing
	T1 lines terminates directly in digital ESS, with no physical
	pairs existing in the CO.  It is a fairly safe assumption that
	an independent operating telephone company saddled with many
	multi-party lines will not have this type of ESS CO.

6.	For options such as fiber optic cable, see (4) and (5) above.

7.	A remote CO suffers from many of the disadvantages of (4) and
	(5) above.  Furthermore, increasing sophistication of subscriber
	line multiplex which terminates directly in a digital switch
	makes these two options almost one and the same.

8.	Rural Radio Service is not applicable to this discussion.

	Unfortunately for those who espouse (myself included) digital
technology and an SLC-96 approach, common old cable pairs are still
often the only practicable solution in a rural environment.

> In article <> (John Higdon) writes:
> What he also neglects to mention is the vastly improved technology for
> delivering dial tone available today. As I mentioned in a recent post,
> Contel delivers literally thousands of private lines to subscribers
> more than twenty miles away from the CO through the same wire plant
> that provided just a few multi-party lines a few years ago.

	I didn't mention it because I was trying to make a simple
point and did not want to confuse the issue.  I have mentioned it now,
though, and as you can see, it did not change my final assessment.

> Growth usually occurs in pockets, and it is a relatively simple (and
> inexpensive) matter to place digital remote offices in the growth
> areas. Contel seems to be able to do it.

	Not true of the serving area in many rural operating telephone

In article  <> (Irving Wolfe) writes:

> I have no contest with Mr. Lippman's construction cost numbers.
> Probably his background is in that field.  It certainly isn't in
> utility finance.

	You're right.  I am an engineer and not much of an accountant.
In fact, I detest accounting and consider it beneath me. :-) However,
at one point in my somewhat varied career I spent quite some time as a
consulting engineer serving the needs of a small retinue of
independent operating telephone companies, none of whom had a degreed
engineer on staff.  During that period, I spent a not insignificant
amount of time helping these telephone companies deal with state
PUC's, banks and the REA with respect to plant expansion.  By the
process of osmosis, if no other means, some aspects of the reality of
utility finance rubbed off on me.

	As some other readers have pointed out, the final assessment
in my original article is not far off the mark.  Times have changed in
the independent telephone industry and are economically difficult for
small independent operating telephone companies.

	Consider a trivial example of say, a hypothetical 1,000 line
independent operating telephone company.  Let's be generous and say
that local exchange service with some message units and toll revenue
(don't forget, some small independent operating telephone companies
are isolated communities and have less toll traffic to the "outside
world" that one may think) averages $ 25.00 per subscriber per month,
which is $ 300.00 per subscriber per year, and $ 300K per year in
revenue.  How far do you think $ 300K per year is going to go
supporting at least 4 full-time employees (minimum by rule of thumb
for this size independent operation), overhead, repair costs,
interest, principal on previous loans, taxes, insurance, etc.?  How
much money do you think will be left over for plant expansion?

	As I said above, this is a tough situation - which is also why
many small independent operating telephone companies can no longer
survive and are bought out by larger conglomerates.  This may be good
news for some, but from my standpoint (admittedly prejudiced) it is
sad because truly independent operating telephone companies of less
than 2,500 lines are becoming an endangered species.

> Ten years is not such a terribly long period for return of investment
> for a phone company, in fact it is probably close to the legal limit
> in most states.

	The problem is, today, many independent operating telephone
companies are facing more *serious* issues involving plant expansion.
Like how long can they keep the SxS or XY going, and how long can they
hold out from PUC pressure to provide subscribers with features most
of us take for granted: like alternate long distance carrier service,
911, DTF coin telephones, DTMF dialing, etc.  We're talking about much
more serious money to replace the whole CO, and the issue of where's
it going to come from.  The days of 1% REA loans are long gone.

In article <> (Peter da Silva) writes:

> But what about new technology? Suppose you put some sort of
> multiplexors on those lines, and ran a bunch of digital voice over the
> major portion of the existing plant? How does that change the
> economics? 

	See beginning of article.

In article <> sheds further
light on a point made in my original article:

> Telcos are not ALLOWED to force people to change from multi-party
> service to (more convenient) more expensive single line service.

In article <> (Marvin Sirbu) writes:

> Patrick argues that after the outside plant is amortized, the
> continuing revenue is "pure profit".  Under Rate of Return regulation,
> the LECs are allowed to earn an authorized percentage (12-16%) on
> investment NET OF DEPRECIATION.  Once the plant has been fully
> depreciated, they are no longer entitled to earn any return on it, and
> telephone rates are adjusted downward accordingly.

	Good point.  Contrary to what some people believe, NO ONE gets
rich running a small independent operating telephone company.

In article <>
(Fred R. Goldstein) expounds upon the frustration shared by *all*
parties in this kind of rural situation:

> This thread began, if you recall, with a comment about rural
> multi-party telephones.  Not city or suburb, but serious boondock
> country where the local telco doesn't even ask you if you want
> single-party lines.
> In such cases, the latest and greatest telco transmission technology
> can help a little, but not a lot.

Larry Lippman @ Recognition Research Corp.  "Have you hugged your cat today?"
UUCP:    {boulder|decvax|rutgers|watmath}!!kitty!larry
TEL: 716/688-1231 || FAX: 716/741-9635      {utzoo|uunet}!/     \aerion!larry

Index Home About Blog