Over at TechDirt, Tom Lee has a sharp critique of Muayyad Al-Chalabi's much-circulated paper (via GigaOm) opposing bandwidth caps. Make sure to read Tom's entire essay, but here's the key take-away:

this whitepaper merely amounts to a complaint that a free lunch is ending. Bandwidth is clearly an increasingly limited resource. And in capitalist societies, money is how we allocate limited resources. The alternate solutions that Al-Chalabi proposes to the carriers on pages 6 and 8 -- like P2P mirrors, improved service and "leveraging... existing relationships with content providers" -- either assume that network improvements are free, would gut network neutrality, or are simply nonsense.

Indeed. But Tom generally agrees that "Comcast's bandwidth cap is a drag" and that "Instead of disconnection, there should be reasonable fees imposed for overages. They should come up with a schedule defining how the cap will increase in the future. And the paper's suggestion of loosened limits during off-peak times is a good one." Well, those are three different things but I generally agree with all of them. Let me just repeat, however, my strong endorsement of the first option -- metering at the margin -- and again highlight the optimal way to do it from an economic perspective. As I noted in one of my many previous articles about metering for bandwidth hogs:

my preferred model [is] what economists call a "Ramsey two-part tariff." A two-part tariff (or price) would involve a flat fee for service up to a certain level and then a per-unit / metered fee over a certain level. I don't know where the demarcation should be in terms of where the flat rate ends and the metering begins; that's for market experimentation to sort out. But the clear advantage of this solution is that it preserves flat-rate, all-you-can-eat pricing for casual to moderate bandwidth users and only resorts to less popular metering pricing strategies when the usage is "excessive," however that is defined.

My former PFF colleague Scott Wallsten penned an outstanding paper on the issue last year entitled, "Managing the Network? Rethink Prices, not Net Neutrality," in which he also endorsed the idea:

Broadband use could similarly be metered. One could imagine simple metered pricing, in which users pay by the bit. Alternatively, providers could develop hybrid plans in which metered pricing begins only after some very high level of usage. In that case, heavy users would pay for the costs they impose on the network rather than being subject to what might otherwise appear to be arbitrary delays in their Internet traffic or threatening letters in their mailboxes. ISPs know how much bandwidth their users use, even if they do not know what content is flowing over the pipes. Implementing new pricing schemes presumably would not be a technical challenge.

I still think this approach deserves a fair hearing, but given the hysteria we have seen over bandwidth cap proposals I suppose that people will just keep looking for a free lunch instead.