Suppose that all over the world, people coordinated so that one in three
households paid ISPs while a neighbor on each side piggybacked (and
perhaps paid the paying househould their one-third share). Do you
really think that would have no effect on the pricing and availability
of internet service?
Are the accounts unlimited downloads, or are are they capped? Does the
ISP charge for excess downloads, or do they slow the connection down to
modem speed? Or even disconnect the account until the end of the month?
Does the ISP offer only a single plan, or are there multiple plans with
different caps? Can people cancel their accounts without notice, or are
they locked in for 12 months? How many ISPs are there? If only one, does
the government enforce laws against anti-monopolistic behaviour, or is it
happy to look the other way? There are far too many variables to give a
definitive answer to your question, but I'll try...
Consider a typical set of neighbours, Fred, Barney and Wilma, all with a
10GB monthly cap, and each use 8GB of that cap in an average month.
Barney gets wi-fi, and leaves it open. Fred and Wilma immediately cancel
their accounts, and piggyback off Barney, and in fact increase their
usage to 10GB because its not costing them anything.
What the ISP sees is that their total usage goes from 24GB used out of
30GB paid for, to 28GB out of 10GB paid for. If they're charging for
excess usage, they'll rub their hands with glee -- excess usage fees tend
to be brutal, and pure profit. No matter how altruistic Barney is, he'll
surely soon upgrade his cap to 30GB (or more).
If the ISP has done their sums right, their profit on a 30GB cap will be
more-or-less equal to 3 x their profit on a 10GB cap -- and very likely
larger. Why? Because of fixed, per account, costs. The ISP's fixed costs
(administrative costs) depend on the number of accounts, which has just
dropped by two thirds. Their variable costs depend on the amount of
downloads, and have increased by one sixth -- but the transmission costs
themselves are quite low. It's not unreasonable to hypothesise that the
decrease in per-account costs more than makes up for the increase in
transmission costs.
Essentially, Barney is acting as a middleman between his neighbours and
the ISP. (The fact that Barney may not collect any money from Fred or
Wilma is irrelevant -- he's just making a monetary loss from the deal.)
Suppliers often, but not always, love middlemen, because they can palm
off the least profitable and most expensive parts of their business to
somebody willing to work for a smaller margin. In this case, the ISP gets
to supply three customers for the administrative and help-desk costs of
supplying one (Barney). It's not unreasonable for this to be a win to the
ISP. Sometimes you get multiple layers of middlemen, e.g. in Australia
it's not unusual to have ISPs like Telstra who deal direct with the end
consumer but also sell bandwidth to smaller ISPs like Internode, who also
sell to the consumer as well as selling bandwidth to tiny ISPs with a few
hundred customers. Would this be viable with thousands of (effectively)
nano-ISPs with two customers each? I don't know, but it could be.