Your Tuesday morning Spontaneous Order
Here's a pretty old post from the blog archives of Geekery Today; it was written about 10 years ago, in 2013, on the World Wide Web.
A marine photographer managed to capture hundreds of wide-eyed fish apparently posing for a picture. Californian photographer and conservationist Octavio Aburto had spent years photographing the school in Cabo Pulmo National Park, Mexico, and had been trying to capture this shot for three years.
Fish are pretty awesome. (So’s technological civilization.)
(Thanks to Cap’n Midori.)
Ever given thought to how tech can give the state more power? Right now we know the state uses the internet and phones to spy on us. Also, technology can strengthen the arm of corporations by making them resistant to things like the knowledge problem (the KP is something advocates of central planning laugh at, since they see tech as being able to solve it). What do you think?
Rad Geek /#
Sure, of course. On the other hand, technology also facilitates new forms of global communication, horizontal connection, grassroots organization, social accountability and decentralized resistance. I can list off examples of each easily enough — No Such Agency, DARPA, CCTV, drones, etc.; but on the other hand Tor, Wikileaks, Arab Spring, Occupy Sandy, CopWatch, etc. The question here is not whether technology helps one side or helps the other (obviously, it helps both) but rather which is going to predominate — the technology of control or the technology of resistance. We are pretty early right now in a really very massive shift and I don’t feel very confident making a lot of predictions about who wins that race; but what I would cautiously suggest is that I don’t think that’s predetermined at all. And while particular technologies can lean towards particular applications, I do not think that in the end this is solely or even primarily determined by the development of the technology itself. What I’m inclined to think is important is the extent to which we take advantage of, and make intelligent use of, the potential that emerging technologies offer us.
There is some talk about this in the comments thread on the “Technological Civilization is Awesome” post.
Advocates of central planning often do take that line, but I think it’s a serious misreading of the present situation — there’s a lot of incumbents right now feeling some pretty severe pressure from low-margin, barely-for-profit or not-for-profit alternatives (think, for example, of what Craigslist is doing to the advertising industry). And in the industries where this is noticeably not happening, it’s very often not because of the incumbents’ use of technology, but rather precisely because the regulatory structure has locked in existing business models, blocking would-be competitors from making full use of potentially disruptive technologies. (Look, for example, at how aggressively the state has moved to block any serious networked innovation in money transfers or microfinance.)
As for the knowledge problem, the idea that sufficiently advanced computers will solve it is of course an old one. But I think one which involves a grave error about the nature of the Knowledge Problem. If we’re supposed to be focusing narrowly on Hayek’s understanding of the problem (so, focusing on the need for price signals in order to aggregate and act on massively decentralized, tacit or localized knowledge), then advanced computers or massive databases can only solve part of that problem. Specifically, they can better centralize previously decentralized information about the quantities or locations of resources. But I don’t think you can solve the Knowledge Problem simply by presuming that all the relevant knowledge can be quantified or expressed in statistical form: since part of the point of the KP has to do with the central importance of transient conditions, tacit knowledge and unquantified or unquantifiable aspects of local cirumstances.
As I understand it, Hayek’s Knowledge Problem is, in a sense, a problem of epistemic overwhelm, but what’s supposed to be overwhelming isn’t just the mass of information already given (of the sort that a really good database or analytics might help you with). Rather, the planner has a prior problem, even before they get to the problem of overwhelming mass: that is, successful trades often make use of knowledge (local know-how and familiarity) that has not been made explicit, and which either cannot be made explicit at all, or else cannot be made explicit without a significant cost in both time and money. And so this involves a great deal of important knowledge that isn’t yet available to a planner’s calculation, and even the process of trying to identify and convert it into quantifiable, statistical information is overwhelming, before we ever get to the problems of scale in information.
If your assumptions about the KP are true, could you explain why it was that the vast majority of state-socialist countries fared much better than capitalist ones for a good portion of the 20th century? For example, North Korea’s economy grew far faster after the war than South Korea’s. China’s development under Mao was much greater than India’s post-colonial development (India had a famine that was arguably much worse than China’s, though it’s not referred to as a “famine” due to ideology).
“But I don’t think you can solve the Knowledge Problem simply by presuming that all the relevant knowledge can be quantified or expressed in statistical form: since part of the point of the KP has to do with the central importance of transient conditions, tacit knowledge and unquantified or unquantifiable aspects of local cirumstances.”
Could you explain this more clearly? It seems like you’re not getting to the meat of the argument and are just saying: “I don’t think it works.” There is a principle which shows that the info collected by several decentralized computers can be just as easily gathered by one single computer.
“that is, successful trades often make use of knowledge (local know-how and familiarity) that has not been made explicit, and which either cannot be made explicit at all, or else cannot be made explicit without a significant cost in both time and money.”
There’s a few assumptions here. First, what exact knowledge you need is based on what task you’re trying to do. You don’t necessarily need full information to get something done efficiently. The fact that big corporations are able to supply according to demand at a low cost in itself shows that you don’t need full knowledge. Things like that aren’t hard to figure out, and the overall gathering of the information isn’t difficult. I’m assuming you’re not talking about the actual production. The idea that there’s tacit knowledge is usually irrelevant. In fact, your entire argument could be used to prove that no producer aside from an individual producing for him/herself would be able to effectively produce. There are information problems, but these are in no way the boogeymen you’re portraying them as.
Second, you don’t need to plan every single little thing. All you need to do is plan a few general things. There are things that are significant and things that aren’t. Most knowledge problems come from decentralization, not centralization, since there can be massive communication problems (decentricism can also significantly raise costs for this reason). One of the reasons large organizations exist is to do away with unnecessary peculiarity. It’s not that the state just decides it wants to create large organizations willy-nilly, it’s that most businesspeople would prefer it since it is far more efficient and money-saving in the long run.
“It’s not that the state it wants its rules enforced ham-handedly by unaccountable violent street gangs, it’s that most cops would prefer it since it is far more efficient in the long run to not have to answer for one’s laziness, cowardice, and cruelty.”
Of course businessmen benefit from a system that centralizes resources under their control. What ‘saves money’ for businesses has no more to do with economic progress or health or whatever than what is convenient for cops has to do with the convenience or safety of society in general.