[LINK] NBN pricing may dull broadband benefits: economist

[Paul Brooks]

On 12/03/2013 9:11 AM, Gordon Keith wrote:
> On Sat, 9 Mar 2013 09:42:28 PM Paul Brooks wrote:
>> He is correct - it doesn't actually cost NBN Co any more, primarily because
>> NBN Co only goes as far as the POI. It costs the backhaul provider more,
>> but that is outside the NBN Co bit. However, it does set up a scale of
>> increasing price for increasing functionality, which sets up a yield curve
>> that can be optimised . If they charged a single price, it would be higher
>> than the current lowest-tier price, which would cause price-sensitive
>> customers to not purchase.  A sliding scale allows them to collect the
>> maximum number of customers, each for as much as they are comfortable to
>> bear.
> But should the sliding scale be on speed or data?
>
> I'm on the cheapest NBN plan, 1Mb/sec 30 GB / month (it's 12 Mb/sec download, 
> but I'm firmly of the opinion that broadband speeds should be quoted as the 
> slowest rather than fastest speed) because 30 GB / month is adequate for my 
> current needs. So long as it is faster than dial up I'm not too fussed about 
> speed.
>
> For users at the lower end I suspect the price sensitivity is on quota not 
> speed. You could have all the advantages of price discrimination without 
> losing then potential benefits of future high speed applications (which will 
> increase customer demand for quota) by having all plans at full speed but have 
> the price scale with quota.

It should be on speed, because the concept of quota is crap, and quotas are steadily
increasing faster than people are using them up to the point where most people are
able to get a service with a quota sufficiently above their normal usage that they
rarely or never hit the ceiling any more.
It almost has to be on speed, because there is precious little else that it could be
tied to!

For the NBN, it *has* to be on speed, because that is the only thing that the NBN
controls. For the retail plans generated by the ISPs and sold to end-users it can also
be on quota, and with a large choice of ISPs and plans available, each end-user can
then choose a plan with maximum speed but a small quota, or a slower speed and a
larger quota, or both or neither, each according to their needs and price sensitivity.

Socially, we do not want to install a system where people are scared to try new
digital technologies, or think they have to limit what they do online, or even have to
understand the intricacies of how much data their particular choice of application and
service is using because they need to keep one eye on a 'data meter' spinning around. 
'Data' is not a finite or limited resource, and in many cases (unlike electricity and
water) the end user is not in control of how much data an application uses, so it is
not equitable or desirable that charging be on a 'per-byte' basis. The ISP industry
quickly worked out that charging on a 'per-kilobyte' data volume charge (as
electricity and water is, and as deRidder apparently wants) was counterproductive and
undesirable, which is how the charging model evolved to quotas and data caps we have now.

Paul.