a la carte TV has arrived

...but probably not in the form or at the price that consumers wish. A recurring complaint has always been, "Why must I pay for 'packages' full of channels that I don't want? I'd rather pay for only the stuff I want."

With iTunes or Amazon, that's possible. Customers truly pay for only stuff of their choosing: no subscription to a package, channel, or even a program. Rather, they buy episode by episode, like buying a periodical off the rack every week or month. This degree of granularity and inconvenience is the opposite extreme of a channel package. It might be just right for some, yet others likely would prefer a moderate option; they'd be willing to pay more in return for a larger conglomerate of programs. Already, specialized "channels" both produce and aggregate programs according to specific interests of the audience. Is there a reason why they couldn't fill similar roles in the domain of digital on-demand entertainment?

Some object to the episodic prices, at two dollars or more, for fresh and in high definition. Of course, the bytes that constitute the episodes are undeniably cheap to transmit/copy, so the "per-unit production" cost is far lesser than the retail price. But the per-unit production cost is always a baseline and never the entirety of the cost to produce and sell. Producers and sellers must cover miscellaneous other costs, e.g. fixed bills like the company payroll or investments like R&D. (Although some products sell at a "loss" in order to stimulate future purchases.) The final price is what the "market will bear"; sellers and buyers settle on an amount acceptable to both. It seems to me that the price level could have a few reasons:

• The most obvious is that the price should be greater than zero to compensate everyone involved in the creation and distribution of the episode. If professionals are to create episodes, then they can expect to trade their work for money. If professionals are to create and manage technology to store and send the bytes, then they can expect to trade their work and fixed-costs for money. Since advertisers aren't contributing (as they typically do during "free" Web streaming), the revenue must come directly from the audience. As with any product, the gross revenue is in effect spread throughout the full "supply chain".
• Prices for fresh and well-encoded digital musical works, the experience of which lasts between three and six minutes, range from $0.69 to $1.25. At its cheapest this is $0.11 per minute of enjoyment ($0.69 divided by six minutes). At that rate, a 40-minute TV episode would be $4.40. This calculation ignores the fact that the TV episode is both aural and visual and therefore each minute is really superior to the song, whether in terms of raw quantity of data or in terms of the subjective impression.
• The episode is bought, not rented. Presumably the price to see an episode only during a limited time frame would be cheaper, which is the rule for the movie "rentals" on these stores. What's being sold is the permission to stream (or download to a scrambled file) the episode for all time; in any case, certainly more than one viewing. While this is different than the traditional meaning of "ownership" (what happens when the seller/provider goes out of business?), it's not too far off. Generally speaking, the usual market alternative that's closer to "true" ownership, a DVD or BD stored on a home shelf, is relatively more cumbersome and expensive (per episode) and also not as timely.
• Massive bundling in TV achieves large-scale economic efficiencies that wouldn't be possible if everything were unbundled. Hence the actual price per channel or program or episode will tend to be correspondingly greater. The supply and demand price curves for each item will vary. Items with greater production costs will cost more. Highly-desired items could cost more (because buyers are willing to pay more). Market competition will be more important than ever, assuming programs can act as substitutes for one another.