Anybody left? (Around here, that probably means "everybody is leftist," as my few conservative readers tend to ignore anything I say is complicated...) These musings are incomplete, at least as of yet; it's hard to have complete musings when events are still evolving. In any event, the general subject — disintermediation — deserves quite a bit more attention than it's going to get otherwise... because, in a bit of infinite reflexiveness that would make any late-1980s artiste squee with fanboy/fangrrl joy, the word changes meaning based upon the immediate personal interests of the speaker, without regard to context.
Although there is a technical, purely economic definition, "disintermediation" most often refers to removing intermediate links between a producer and an end-user of a commodity; in contemporary discourse, that is usually an information-based commodity. For example, the moronic "long-tail theory" espoused by ignoramuses like Chris Anderson at Wired (see the second-to-last sausage on this platter for a beginning point) asserts that all artists would be better off by following a model that applies to at most a little over 2% of all artists, in reality applies to far less than that... and is balanced by corresponding abject economic failures. In this model, artists get rid of everyone between them and the end-users: No recording studios, record labels, concert promoters, and/or record stores; no editors, cover artists, printing plants, warehouse/distributors, or retail outlets. The underlying assumption here — and it's easily refuted by saying "Van Gogh" — is that the intermediate roles require nothing but the will to fulfill them.
Nothing. No expenditure of resources; no independent motivation; no opportunity costs; no product improvement prior to market through feedback.
And so, we end up comparing the Macmillan/Amazon kerfluffle with the Google Book Search settlement. Wait a minute... how did he end up there? They're both about just how much intermediation we're willing to tolerate... and they expose substantial doublethink in redefining terms for the moment.
On the one hand, a not-insignificant portion of (potential, mostly) e-book consumers (who won't buy from Amazon at any price... because they're pirates) applaud Amazon for standing up to the big bad bully Macmillan that is trying to raise the prices of some e-books. That crowd is rallying around "$9.99 or fight" (echoes of "54-40 or fight" and its attendant hypocrisy purely intended), accusing Macmillan of undermining truth, justice, apple pie, and free e-books, and often invoking the disintermediation meme as a purportedly nonnormative justification. The real problem is that this actually reinforces inserting an additional intermediary in the long term: If the model Amazon has demanded (Amazon sets all pricing and compensation terms) becomes dominant, it entrenches Amazon as an additional intermediary and discourages publishers from developing direct-sales systems that — if one accepts the pollyannish enthusiasm of the long-tail advocates — will eventually be adopted directly by writers, thereby cutting out both stages of intermediation. Hugs and puppies for everyone, right?
Conversely, one could look at the problem through the publisher's lens for a moment, and come to equally disturbing conclusions about antitrust and publishing (in particular, the commercial-fiction segment). From this perspective, Amazon is attempting to use oligopsony power to force a particular set of economic terms upon the publishers as the sellers of the goods in question, in a market that already has nearly crippling customs in favor of the downstream actors. From the returns system to net-90 and net-180 payments, from paid-for endcaps to complete noncommunication between chain buyers and floor salescreatures, there is an almost incomprehensible set of barriers — worse even than in automobile sales (and that's really saying something). And here is one of those oligopsonists, demanding terms that further restrict the publisher's flexibility and ability to adapt to changing conditions... No hugs or puppies for anyone.
It gets worse when one looks at how the Google Book Search settlement would operate... because functionally it represents massive new intermediation, in multiple steps. The first step is the most obvious: Google itself as a 500kg gorilla sitting on top of the distribution of out-of-print works in electronic form. And it won't be free: Google is a business, and is going to find a way to stick someone else with the bill — ads (that consumers pay for in the increased price of the advertised goods and services), access fees from libraries (that consumers pay for through increased taxes and user fees), whatever. A little profit would be nice, too, eh? The second new intermediation step, though, is less obvious: The "trust" to handle orphan works, which will (based on historical experience) suck any revenue stream dry with "administrative costs", at least for the first few years... and also represents a disturbing potential for censorship.
What these two contemporary examples tell us about disintermediation is that we're not really talking about disintermediation. We are, instead, talking about rents extracted as a price for adding value. This is, indeed, where it becomes apparent that both Amazon and the GBS settlement mechanisms deserve further scrutiny as potentially wasteful intermediaries... and the publishers deserve scrutiny, but not as intermediaries. Neither Amazon nor the GBS settlement system concerns the actual addition of value to a product or service; instead, they represent purported marketplace efficiencies. In the strict, neoclassical view that economics is only about the efficient allocation of goods and services while minimizing transaction costs, a rent extracted for facilitating a marketplace is justified if, and only if, that rent is less than the increased efficiency actually achieved through that facilitation. Anyone who knows anything whatsoever about the arts — let alone publishing — can recognize that there are multiple, unmeasurable assumptions in there, and that instead we just assume that centralized distribution is more efficient in the arts because it is often more efficient for fungible commodities.
So, in the end, "disintermediation" turns out to be a rather useless meme in this context; we are, instead, considering the identity and nature of the intermediaries we are willing to accept. And that is not a nonnormative argument... which, in turn, demonstrates that — even without acknowledging that both the Amazon/Macmillan and GBS settlement transaction frames are between intermediaries and do not include either the actual producers or the actual consumers — economics, and in particular neoclassical economics, cannot help us find an answer. An economic analysis only shows us exactly what kind of tango we're really dancing.