Back to Dan Kegel's ISDN page

ISDN Tariff Filings & News

Updated Dec 96 (Maryland page updated).

Other resources

Date: Thu, 29 Feb 1996 18:56:03 -0500 (EST) 
From: James Love 
To: Multiple recipients of list 

ISDN tariff disputes

A number of states are currently embroiled in controveries over ISDN
pricing.  A few days ago the Virginia Commission ordered an investigation
into Bell Atlantic's tariffs.  California is involved in disputes over ISDN
pricing with PacBell (and GTE).  Washington State, Utah, New Mexico,
Arizona and other states are engaged in disputes over US West's ISDN
tariffs, and there is much more.  We hope to have a more complete
discussion of this in a week or so.

Tariff Info by Provider and Region

Not Yet Classified

From: (Fred R. Goldstein)
Newsgroups: dc.general,comp.dcom.isdn
Subject: Re: residential ISDN rates
Date: 4 Jan 1996 23:03:04 GMT

In article <4chev8$>, says...
>In <4ch1q5$> Dr. Roger M. Firestone said:
>> ... The Bell System has a very very small incremental cost
>> to carry a phone call for an additional minute.  ...
>  I wonder about that.  In the very short run clearly the cost of
>electricity flowing is miniscule, but to prevent busy signals they
>need to have a large enough plant.  If demand is growing, there will
>be a series of expansions over time.  To the extent an incremental
>minute brings them closer to full capacity, and thus moves their
>investment schedule nearer in time, its full cost must include its
>effect on the discounted present value of that future outlay stream.
>And to the extent that it doesn't cause them to move their
>investments forward (because of lead times or whatever), then its
>cost must include the additional probability of other folks getting
>busy signals.  Or not getting ISDN for months because the CO is
>'full'. :(

Such cost studies have been done.  Over and over.

The *incremental cost* of *local usage* usually comes out to be around
half a penny a minute, all told.  That includes the weighted cost of
having to add more capacity, etc.,etc.  Anything over that is profit.
When telcos exchange local traffic with one another, the accounting
is often in that range.  Some wide-ranging extended local calls cost
more, but even long intra-LATA calls typically cost telco no more
than 3c/minute, under fully distributed cost accounting.

Measured local service is almost always MUCH higher than cost, and thus
generates a cross-subsidy to pay for other things -- telcos are very
inefficient in a lot of regards, and tend to fail at many ventures.
But local usage (ISDN or POTS) is pretty much a guaranteed revenue
stream, so they try to milk it for many times its cost.  Only a monopoly
can do that for long.  Flat rate is actually closer to cost, since more
than 3/4 of the telcos total cost structure is totally insensitive to
Fred R. Goldstein   k1io   +1 617 873 3850
Opinions are mine alone; sharing requires permission.

From: (Doug Humphrey) Newsgroups: dc.general,comp.dcom.isdn Subject: Re: residential ISDN rates Date: 5 Jan 1996 11:25:16 -0500 In article <4chev8$>, wrote: >In <4ch1q5$> Dr. Roger M. Firestone said: >> ... The Bell System has a very very small incremental cost >> to carry a phone call for an additional minute. ... > I wonder about that. I wonder about it too. The telephone company has a very LARGE incremental cost to carry the call, because their facilities are shared facilities in many parts of the call path. To illustrate: Your local loop - that is a sunk cost, only you can use it, when you are not using it, it is not in use. Result - they don't care how much you use the local loop. This is, speaking strictly of the average case and not the occaisional exception, true for the wire from your location to the telephone company switch (please, if you know of some really strange exception, keep it to your self, we are talking about the average case here, and should not get sidetracked on esoterica that doesn't matter) The telephone company switch - assume that we are dealing with modern switches, #5 ESS, DMS100/250, Erikson, etc. These switches are expandable, and the expansions are very expensive. The switch has processors called SwitchMods (SM for short) which are the basic element of switching. A given SM can handle X number of calls IN PROGRESS. This means that one "wall clock minute" of time is X minutes of switching capability. There is a hard limit to the number of calls than an SM can handle at once (varies with the make and model involved) but that means that you have to add more SM's (expensive) once one is "full" to keep from 1) failing to provide dialtone to a customer when they pick up the phone and want to make a call, and/or 2) failing to be able to process an incoming call that is coming in from the interoffice net and needs to be terminates on a local loop. While the initial installation of this facility is a "sunk cost" the expansion of the facilities past their original design parameters is NOT a "sunk cost" - when SM time is not being used by you, it can be used by someone else, as it is a shared resource. As an example, if you have a full time SLIP/PPP connection dialed up and never hang up the phone, then you are effectively reducing the capacity of an SM by 1/X where X is the number of in progress calls it can handle. If a lot of people do it, it begins to effect the traffic flow through the office, and the telco has to expand facilities to stay within the design parameters, or else they will begin to degrade the quality of service that they provide. Interoffice Network - If the call destination is not in the same CO (Central Office) as the originator, then the call has to travel over the Interoffice Network. This network is of finite size, and the calls are all carried as full channel (56 or 64k) DS0's, one per voice call. There are 24 of these in a T1 line, 28 T1's to a T3, 3 T3's on an OC3 (optical carrier) and on up the line. These facilities are very VERY expensive to expand, not only because there is fiber to be pulled, but because the electronics on the ends of the glass (Sonet based stuff) is expensive. As with the SM above, these are NOT sunk costs, except for initial capabilities, because if you place a call and never hang up, one of those DS0 channels is in use as long as you are on the call. If you were not using it, it would be available for others to use. Toll Access Facilities - there is a whole class of facilities which are used to access Intra and Inter-LATA toll facilities. This is the same basic case as the Interoffice network, and toll tandems are much like SM's - they are not sunk costs, they are facilities which are shared by nature, so use of those facilities in excess of the average use which was projected in the design of the facilities causes additional facilities to be constructed, as great costs. So, I would have to conclude that the telephone company has very great incremental costs, and that these are what need to be taken into account. ISDN is treated exactly the same as a voice call (or, in the case of 128k ISDN, 2 voice calls) for the purposes of this discussion - it loads up SM's, it eats interoffice bandwidth, and it has a very high incremental cost. Doug
From: (Matthew T. Russotto) Newsgroups: dc.general,comp.dcom.isdn Subject: Re: residential ISDN rates Date: 6 Jan 1996 10:15:57 -0500 Organization: FishNet In article <4cjjdc$>, Doug Humphrey wrote: }So, I would have to conclude that the telephone company has very great }incremental costs, and that these are what need to be taken into account. }ISDN is treated exactly the same as a voice call (or, in the case of 128k }ISDN, 2 voice calls) for the purposes of this discussion - it loads up }SM's, it eats interoffice bandwidth, and it has a very high incremental cost. Incremental costs at non-peak times are next to nothing. Incremental costs at peak times (which is what the company must size its network for) are significant, but without the numbers (which you have not provided), "very great" seems to be an unjustified statement. -- Matthew T. Russotto "Extremism in defense of liberty is no vice, and moderation in pursuit of justice is no virtue."
From: (John Higdon) Newsgroups: ba.internet,comp.dcom.isdn,comp.sys.mac.comm Subject: Re: ISDN - Whistling in the Wind Date: Sat, 06 Jan 1996 14:46:42 -0800 In article <4ck8nv$>, (John Navas) wrote: >With all due respect, ISDN is not "designed for dialup, intermittant >service" -- it's old technology designed as a cheap way to significantly >increase the capacity of the *existing* copper local loops (albeit much less >than would be possible with newer technologies like ADSL and SDSL). In >return for a modest investment, the telco gets substantially increased >revenue from the same utilization of the same physical plant, even without >further rate increases, which is by any measure a good deal for the telco. I'm afraid I disagree. There are some significant constraints on the availability of ISDN with regard to plant requirements. There are distance from the CO considerations, pair branching and multiples, as well as the obvious switch and software package considerations. ISDN is definitely not something that is just slapped into place on top of existing plant with trivial effort. >Nevertheless, even though subscribers are bearing most of the capital cost >(i.e., on-premises equipment), the telcos are trying to get subscribers to >bear even more of the capital cost through onerous "installation" fees. What do you suppose the installation fees might be from a Pac*Bell competitor who has 100% of his new equipment capital cost to recover? >That is of course a great deal for telco shareholders, who would normally be >expected to be responsible for financing capital investments. That the >telcos get away with this is a function of both the lack of competition and >the lack of economic knowhow in regulatory agencies. Again, I disagree. Agreed that the shareholders should be financing capital improvement and agreed that they are not doing so here, but it was the general ratepayers who funded Pac*Bell's ability to offer high-end services--not just the ISDN customers. >Finally, any monopoly (or monopoly wannabe) knows that metering is the best >way to extract monopoly profits (because once the more or less fixed cost is >recovered, the rest is essentially all margin). The problems with metering >are that (1) it greatly increases costs (measuring, billing, collections, >accounting, etc.), and (2) distorts the economics to favor low demand uses >(which inevitably favors business over personal use, since the per minute >value is higher to the business than it is to the individual, and inhibits >the spread of technology). My own feeling about measured service, after weighing all the arguments, is that business "peak" (8-5) periods should be measured and all other times should not. One of the main arguments in favor of measured service is that capacity is dictated by usage. Since it is the business-day peak demand that determines required capacity, let those users bear the cost (or use the charge to control demand). At all other times, the incremental cost for supplying connections is, as you imply, down in the noise. >Nevertheless, the >proper way to solve that problem is to greatly broaden the base of ISDN, >with packages that are financially attractive to the average consumer, >rather than to reduce demand through higher prices. A increase in metering >will only serve to narrow the potential base of ISDN. Exactly what PB is after. Demand for ISDN is well beyond projections. The new pricing structures will soften the demand to more comfortable levels for Pacific Bell. It isn't what we customers like, but it is the marketplace at work. -- John Higdon | P.O. Box 7648 | +1 408 264 4115 | FAX: | San Jose, CA 95150 | +1 500 FOR-A-MOO |+1 408 264 4407 | |
From: (Doug Humphrey) Newsgroups: dc.general,comp.dcom.isdn Subject: Re: residential ISDN rates Date: 6 Jan 1996 12:22:52 -0500 >The moment Pacific Bell starts charging for packet flow on my frame relay >connection, alternatives will be employed. I agree that nailed up ISDN >connections are a misuse of tariff and technology, but many telcos offer >little else. Pac*Bell at least has a robust FR offering. You are bang on with this John; What *I* want to see is a comparison of telco actual costs for a nailed up ISDN vs. a 56k Frame or even a T1 frame connection - knowning the costs of telco facilities, ISDN seems to me to be using the same amount of wire/interoffice net and much more expensive hardware and engineering than the FR stuff is. So, how do they justify the costs that they are charging for FR? A residential FR would be a GREAT thing for a tariff; 24x7 connection, faster and more reliable than modems, no involvement of the telco voice switching fabric, and scalable easily to T1 levels. Hmmm... Maybe time to take to the PUC again. Doug Actually, FR takes only a fraction of the Interoffice if there is a FR switch in the serviing office; it only uses the same amount of bandwidth as ISDN if the connection is being backhauled to an FR switch in another office.
From: (Doug Humphrey) Newsgroups: dc.general,comp.dcom.isdn Subject: Re: residential ISDN rates Date: 6 Jan 1996 12:40:57 -0500 >Depends on your definition of large. A mil here, a mil there, and >you still have a ways to go to get to a penny. At a cost of even a half cent per min for a nailed up 24 call, we are looking at a telco COST of $7.20 a day, right? And people want unlimited ISDN for $30 a month or less. Let's see: $7.20 x 30 = $216/month TELCO COST This makes my point exactly. ISDN is very expensive in it's incremental componant. "a penny" doesn't seem to be too large to you, but do the math and you will see that this just does not scale. IF your figure of half a cent a minute is anywhere near close, that $30 a month will only buy the customer 100 hours a month of connection, at Telco breakeven. Three hours a day or so? I have a feeling that a SUBSTANTIAL percentage of the users of this service, if they are using it for INternet access (and let's be real, ISDN has not really take off for any other application FOR THE MASSES) are going to be over this number. AND this is supposed to be the telco cost; they are allowed by LAW to collect a profit above and beyond... >I've seen some numbers, and know people who've studied them in detail. >When you add them ALL UP, they come out to a much, much lower figure >than you'd imagine. They're higher than a milli-lire per minute, and >lower than a dollar. The actuals TOTAL usually around half a penny a >minute, for a short-haul (most areas all you get) "local" call. As I said above, half a penny a minute is big, big trouble. It is the absolute death of a flat rate ISDN service in the long run, and it proves very well the statement that ISDN does not scale at the levels that it will have to in order to be a success. Doug
Date: Wed, 17 Jan 1996 17:24:10 -0500 (EST) From: "James E. Leinweber" To: Multiple recipients of list Subject: cost of residential ISDN The comparision appears to be accurate for Ameritech/Wisconsin. I just checked with them, and the rates are indeed $113.05 installation, plus $30.90 per month flat rate, plus $5/month local access fee, plus $.06/call (same as analog voice service) for 2B+0D residential ISDN. An ISDN line is priced at roughly twice the rate of a voice line, which is plausible. The (proposed?) Bell Atlantic rate James Love analyzed at $1176/month for 24 hour a day nailed up service is totally outrageous. I've got a 5 mile T1 line at work which costs $567/month full time, and that's probably an excessive rate. Figuring a pair of ISDN B channels as 1/12 of a T1 line, anything over $47/month for flat rate unlimited service is monopoly profiteering. Ameritech is "off the hook", at least so far :-) -- James Leinweber State Laboratory of Hygiene/University of Wisconsin Madison WI, USA fullfeed!stovall!jiml +1 608 262 0736
Date: Thu, 18 Jan 1996 16:53:09 -0500 (EST) From: James Love To: Multiple recipients of list Subject: BA, ISDN and SS7 deployment Taking David Lesher's advice, we checked with Fred Goldstein of BBN about the issue of whether or not BA installed SS7 switches in order to provide residential ISDN. This was his reply. jamie ---------- Forwarded message ---------- Date: Thu, 18 Jan 1996 14:07:03 -0500 From: Fred R. Goldstein Subject: Re: Latest BA Bell Juice Bell Atlantic is taking extreme liberties with the truth if they attribute SS7 deployment to ISDN, residential or otherwise. SS7 is needed for revenue-generating CLASS features such as Caller ID and City Wide Centrex. BA has t he highest Centrex penetration in the world. Over 2M lines, many in DC (GSA). They use SS7 for everything, including faster call completion. They had SS7 way before NYNEX; I first noticed it when calling from my mother's place in NJ to a distant part of that LATA, around 1991, and the call went thru in under 1 second. NYNEX was a LOT slower! $32/month is a good rate, too. Even with 2 EUCLs atop it. ___ Fred R. Goldstein BBN Corp. Cambridge MA USA +1 617 873 3850
Subject: RTP, NC ISDN users' group Newsgroups: comp.dcom.isdn From: Date: Sat, 27 Jan 96 21:00:46 EST BR>Come on, folks. Also, I don't know much about this kind BR>of stuff, but could some one tell me how I may begin to BR>petition the PUC for Bellsouth to lower the monthly rate BR>for ISDN? Or if that last sentence didn't make any sense BR>please tell me what I should do. BR>-- BR>Thanks, BR>Brent J. Miller BR>IBM, Inc. BR>Systems Analyst, ISDN Specialist Brent; I can't speak for Bellsouth, but several years ago I represented my company in a Radio Common Carriers association that constantly did battle with NYNEX over interconnect rates. The most helpful advice I could give you is to form a group with common interests and get a good telecomm attorney to help your group. I can give you the names of several via email if you desire. We found the PUC in Mass. of limited assistance; however, the FCC forced NYNEX to the negotiation table. After 2 years, the RCC group finally got some VERY good rates! Good Luck! John -- =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=- This message brought to you by the RadComm BBS, Lynn, MA (617)477-0852) -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
Date: Tue, 30 Jan 1996 12:15:28 -0500 (EST) From: James Love To: Multiple recipients of list Subject: ISDN Pricing Workshop on Feb. 28 We have reserved a room at the Hotel Washington on February 28, 1996, for a workshop on ISDN pricing issues. The very same week the National Association of Regulatory Utility Commissioners (NARUC) will be having their winter meeting in DC, at the same hotel. I am working on a program, and suggstions are welcome, as are joint sponors of the workshop. We are considering a program that addresses policy issues (why should Public Service Commission's care about the issue, what can broader ISDN deployment do for consumers and information providers... etc), and discussions about costs and technology... I'll be posting follow up information on ISDN@ESSENTIAL.ORG, the isdn pricing discussion list (subscriptions from Let me know if you have suggestions regarding speakers, program, or sponors... jamie ---------------------------------------------------------------------- James Love, P.O. Box 19367, Washington, DC 20036; v. 202/387-8030; f. 202/234-5176 Consumer Project on Technology; Taxpayer Assets Project;