Re: cost of residential ISDN

[hes@unity.ncsu.edu]

  Jamie writes:
  >I would welcome pointers to publicly available studies of residential 
  >ISDN costs.  From what I've been told lately, one can think of residential 
  >ISDN costs as fixed costs (F), and variable costs (VC), which relate to 
  >demands on peak switching capacity (s) and interoffice trunkage (t).
  
    The usual cost analysis has to have a few more portions - because
  "fixed" isn't really totally fixed, it can be fixed with respect to the
  usage by an existing customer, but it usually has a stair-step graph for
  cost as new customers are added.  Variable usually is really variable - 
  e.g. using up supplies or power when doing something.
  
    The physical facilities (lines, switches, buildings) are usually 
  considered to be "fixed cost" items - in that they cost essentially the
  same (fixed) amount whether or not they are used.  So the cost is fixed
  with respect to the amount of usage - until the capacity is exhausted
  and a new item (new switch or upgrade, new line(s), ...) is needed.
  This particularly happens when new users appear or new lines are
  requested.  That's not usually considered to be "variable cost" - 
  simply to avoid the undesireable ideas that the variable cost for most 
  use/users is zero, and then it's $1million for the next use.  So to be
  precise we really need to talk about incremental (marginal) cost for
  extra use by a user (e.g. when I make one more phone call from home), or
  for extra users (e.g. when I add a line to my home, or a new home is
  connected.)
  
    What's variable cost in a telco setting?  Certainly the time the telco
  staff spends talking on the phone to help customers, the extra
  electrical power needed for equipment which is actively working instead
  of being in idle mode - and some other small items.
  
  >I was also recently told of a cost study of residential POTS which 
  >indicated that fixed costs were about 93 percent of the total, and 
  >variable costs were about 7 percent.  If this is right (and maybe a 
  >different number is better)... 
  
    This sounds reasonable.  But
  
  >then every $1 (per month) in the POTS 
  >tariff would cover a 100 percent increase peak load capacity of the 
  >service.  
  
    I just don't see how this follows.  Where does it come from?
  
  >This would suggest that the Consumer Federation of 
  >America's estimate of $4 per month as the incremental cost of upgrading 
  >residential POTS to ISDN is quite reasonable. 
  
    I also don't see how this follows.  (It may be right - I don't see how
  one can deduce this from knowing that POTS costs are 93% fixed.)
  
    E.g. 2 scenarios for my home -  served by unshared copper pairs from a
  CO somewhere not too close.  (Note that these pairs represent a "fixed
  cost" with respect to my usage.)
  
  I.  These are nice pairs, and they go to a CO < 18,000' away, and will 
  carry ISDN signals with no problems.  Incremental cost of upgrade $0/mo.
  
  II. These are crummy, corroded pairs with bad splices, moisture, bridge
  taps, and termites - but they work (barely) for POTS, and go to a CO
  >18,000' away.  To upgrade to carry ISDN signals will take complete
  replacement plus added/powered electronics to boost the distance.
  Incremental cost of upgrade $much/mo.
  
    My point is that in both cases the costs are fixed - and the upgrade
  cost has nothing to do with that at all.  We could similarly add to 
  these scenarios differences in switches regarding cost of upgrade to
  handle ISDN, etc. - and still have the %fixed costs provide no useful 
  information about how much upgrading will cost.
  
    Considering that ISDN was designed to require the minimum possible
  upgrading of the telco outside plant, and that for many years the
  swtiches have been designed to handle ISDN well and without penalties -
  I wouldn't at all be surprised to find out that the "incremental cost of
  upgrading residential POTS to ISDN is quite reasonable."  I just don't
  see that it can be deduced from the numbers given.
  -- 
  --henry schaffer
  hes@ncsu.edu