The problem with this is that the actual transit cost per GB never seems to come close to what the ISPs charge. $1.20 per GB could very well be 100 times what the ISP ends up paying to actually move the GB. The basic problem is that the up-front infrastructure and maintenance costs are expensive, but the transit is extremely cheap. So, unless the ISP charges a base minimum amount, a heavy user just ends up heavily subsidizing the infrastructure and maintenance costs of the light users.
Not to mention that charging by byte is a great way to stifle innovation on the web. Basically, I would disagree with your assessment.
Yes, you are correct about minimum account fees being required to avoid a subsidy in the other direction (large->small). NZ packages typically work out to about US$24 + $US1.20/GB, although the per GB price is dropping. We're getting fiber in the next couple of years, and at least one ISP is already offering TB/mo data bundles for US$144 (incl 100/50 fiber).
I don't see how it stifles innovation at all. It does stop a cross subsidy from the ISP to the company attempting to implement a new, even higher, usage service.
I appreciate the reply and further details on these NZ plans. Given the rather high cost of $1.20/GB, I naively assumed that these plans did not actually also have a minimum account fee. I would still contend that $1.20/GB is ridiculously high for home broadband, barring special circumstances (remote areas, cell network, etc).
As for stifling innovation, consider what the internet would look like if we were all still on dial-up. Greater bandwidth continues to open up new innovations; customers struggling with low bandwidth or expensive per GB pricing schemes hinder that innovation.
I can see that. I just look at the NZ market and see competition working. :)
You're right, $1.20 feels expensive and we all grumble about it. Even so, I did some math and found out it was cheaper to stream video at $1.20/GB than it was to pay for cable tv, so I cut the cable and upped my data bundle. :) It's all relative.
I'd argue that it's only when there's a single monopoly provider does pricing start to stifle innovation. The NZ ISP market is split into wholesale/retail, with an unbundled local loop and a regulated wholesale price. ISPs can purchase DSL connections from the wholesaler, or pay for access to the cabinet to install their own equipment.
Even when NZ was analog they had innovation. The biggest one was the creation of "Free" ISPs, where they lived off of the termination fees that they were able to charge the originating carrier - much the same as free conference calling and international long distance services out of Ohio. That lasted about as long too. :)
NZ has tried unmetered service several times, with really bad results. Throttling, low bandwidth, ugly. High traffic users would shift to the unmetered plan and overwhelm the provisioning ratios. You ended up with a service that only bulk users (low throughput, high volume) would want to use.
However, since unbundling (2006), we've had pretty vibrant retail ISP competition, which goes through ebbs and flows of price competition. The shift to fiber is currently causing lots of competition. :) We've got 3 mobile carriers and 7 major ISPs. The ISPs all offer fixed phone service as well, some on their own equipment, some from the wholesaler.
The carriers have learned that the market isn't going to pay extra for fiber, but it is expecting more service and bundled data. Some are currently grumbling about that, but their competitors are jumping on the treadmill so everyone has to. We're starting to see fiber accounts with 1TB for US$115, which is US$0.12c/GB, a much better price.
I look at my connection and I've got a 100/10 link, and I can pull 30-50mbps of international traffic out of it during busy hour. I can get the full 100mbps to the ISP's servers - the local Steam mirror can be wicked fast. I attribute this to the carrier perceiving me as a source of revenue to be encouraged instead of a cost to be driven out.
Not to mention that charging by byte is a great way to stifle innovation on the web. Basically, I would disagree with your assessment.