There ‘s a useful post over on GigaOm about Cloudonomics – it has 10 laws, which are (in summary)

Cloudonomics Law #1: Utility services cost less even though they cost more.
Although utilities cost more when they are used, they cost nothing when they are not. Consequently, customers save money by replacing fixed infrastructure with clouds when workloads are spiky.

Cloudonomics Law #2: On-demand trumps forecasting.
The ability to rapidly provision capacity means that any unexpected demand can be serviced, and the revenue associated with it captured.

Cloudonomics Law #3: The peak of the sum is never greater than the sum of the peaks.

Cloudonomics Law #4: Aggregate demand is smoother than individual.
Aggregating demand from multiple customers tends to smooth out variation.

Cloudonomics Law #5: Average unit costs are reduced by distributing fixed costs over more units of output.

Cloudonomics Law #6: Superiority in numbers is the most important factor in the result of a combat (Clausewitz).
The classic military strategist Carl von Clausewitz argued that, above all, numerical superiority was key to winning [DDos] battles.

Cloudonomics Law #7: Space-time is a continuum (Einstein/Minkowski)
A real-time enterprise derives competitive advantage from responding to changing business conditions and opportunities faster than the competition.

Cloudonomics Law #8: Dispersion is the inverse square of latency.
Reduced latency — the delay between making a request and getting a response — is increasingly essential to delivering a range of services….However, to cut latency in half requires not twice as many nodes, but four times.

Cloudonomics Law #9: Don’t put all your eggs in one basket.
The reliability of a system with n redundant components, each with reliability r, is 1-(1-r)n. So if the reliability of a single data center is 99 percent, two data centers provide four nines (99.99 percent) and three data centers provide six nines (99.9999 percent).

Cloudonomics Law #10: An object at rest tends to stay at rest (Newton).
A data center is a very, very large object. While theoretically, any company can site data centers in globally optimal locations that are located on a core network backbone with cheap access to power, cooling and acreage, few do….. A cloud service provider can locate greenfield sites optimally.

I think this is a very good discussion on the economics of the Web Hosting side of the equation and this is exactly the same arguments that Web Hosters like Exodus (remember them) used 10 years ago, and still do today. But, I would add a rider – this discussion covers the big datacentre (even better a number of big datacentres networked together) economics, but that is not the entire digital logistics supply chain for Clouds, Grids and other Webservice systems

The point I’d add here about Cloud reliability is that the datacentre reliability got there via redundancy (3 datacentres), but you need to match that in the rest of the system. You also have to have reliable connectivity, as you are limited by the reliability of the weakest link – so in this case we would need 3 connections (each of a separate frontbone and a backbone, which have to together equal 99% reliability) of equal reliability to keep the entire system operating at that quoted level of reliability. But, how many webservice clients will have 3 ISP’s coming into their offices, each on different pipes, from different directions (in case the Utility Co pulls the street up and cuts all 3 pipes).