Scott versus Hayek

lumping concrete things that are differentiated in important ways under abstract headings

As a devil’s advocate, I think a Hayekian would respond that the market would subdivide that category if enough people had a strong enough interest in doing so, and so different things are lumped together in the market only when not enough people care about the differences to justify the transaction costs associated with preserving them.

Thus, even the market’s decision to destroy information reflects a tacit judgment of the uselessness of that information.

“But Hayek is remarkably incurious about the actual social processes through which markets work, and in particular the forms of standardization that are necessary to make long-distance trade work.”

What about Law, Legislation and Liberty, Volume I? You cite a very important article … an article where you can’t make every point in the world.

“But global markets do not rely on tacit knowledge. They rely on standardization.”

Why oppose these two things? Even if standardization is a process distinct from the coordination and aggregation of tacit knowledge (is it?), I can’t see a reason to think global markets don’t rely on both.

“Far from communicating tacit knowledge, the price system (and the codified standards that underlie it) destroys it systematically.”

Not all tacit knowledge is created equal. Hayek thinks that markets aggregate information needs to coordinate; if standardization is what is needed to coordinate, then why is it a problem not to communicate unstandardized information? Hayek knew that prices were crude summaries of information; that’s the point, really, right?

“If markets both distort production (by pushing towards products that can be standardized) and destroy knowledge (by lumping concrete things that are differentiated in important ways under abstract headings), then the Hayekian case for markets is at least qualified.”

Why think that markets distort production by generating standardized products? I’m confused.

Few academic classical liberals think that markets don’t rely on institutional frameworks that aren’t always constructed through spontaneous order and profit-maximizing processes. Vernon Smith’s work in experimental economics has made clear that norms of non-maximization are required to create expectations needed to make markets efficient. Lin Ostrum has taught us a lot about how the affirmation of moral norms helps to produce a social order. And I think Hayek understood this; It’s really clear in LLL. Heck, even Adam Smith, Hume and Mill understood this. Even lots of contemporary classical liberal philosophers emphasize this point, including Loren Lomasky, David Schmidtz, Gerald Gaus, Chandran Kukathas and Jacob Levy. Which economists are you talking about? Most of the GMU-econoblogosphere is highly conversant in the material you claim Hayekians lack.

In fact, which classical liberals ever made the claim you think Hayekians are committed to? Who are these nameless Hayekians? Why don’t you actually cite a classical liberal that believes what you say Hayekians miss? Could you? Because the more I think about this the more I have no idea who you’re talking about. It could just be my ignorance.

A fascinating topic. I’m glad to see the quote from Scott on the “level of simplification and tunnel vision” and standardization driven by industrial capitalism — I regretted the omission of that issue from Seeing Like a State.

All of this raises the question “What forms of large scale equilibrium preserve or enhance local, tacit knowledge?” If to change we have to choose a new vision, not just reject the old one. And I think this is a very difficult but also very fertile and helpful question to address.

I do think we have to respect the need for rough equilibrium (in the rational expectations sense, if not with the rational expectations background assumptions). Our models have to explain the self-maintenance of a relatively stable order within which individuals can make choices.

Unfortunately I don’t have a proposal to answer my own question. I’m not yet even at the point where I can articulate a clear list of the properties a model would have to have, to generate global equilibrium without forcing standardization or otherwise destroying local fitness.

Maybe ecologies are worthwhile examples. They seem to reach a rough equilibrium while not imposing anything I can interpret as standardization. But I don’t know of any way to analytically model that equilibration process.

One thing I have noticed: Sufficiently fine grained standards don’t interfere with the deployment of local, tacit knowledge. For example the standards that make the internet work are so fine grained that they don’t interfere with evolution at the level of content and applications. Similarly the standards for voice compression and transmission in mobile phone protocols are fine grained enough to let people talk however they way. Without going down a technical rathole, this is not a trivial design choice, many protocols try to structure things more “semantically” and cause all the problems that one would expect given this discussion, eventually failing due to mismatch with evolving requirements.

“Destroying knowledge” is a funny phrase. The price of turnips in your example is certainly a “lossy” form of compression, but presumably the turnip farmers retain the additional knowledge. It’s not communicated to buyers, because they don’t care about the other stuff. The distortion could possibly make a particular variety of turnip standardized, and then knowledge of the non-standard types would be devalued. So it’s hard to separate the distortion & destruction.

I recall in Scott’s book the issue of tacit knowledge is sometimes related to an insiders vs outsiders dynamic. A “dynamist” described/lauded by Virginia Postrel would celebrate the shaking up of exclusive, entrenched networks. David Brin, who describes himself as a left libertarian, made a similar argument for the merits of progressive sci-fi vs reactionary fantasy literature (always with “the chosen one” from birth, often the member of an elite caste).

James Fallows had an article in the Atlantic on business in China. Their industries sell globally to big established American corporations, but the Chinese firms themselves (in the studied city) tend to be very small and unheard of. There was an Irishman who had been living there for a while and spoke no Chinese, but was reputed to have a great advantage from “tacit knowledge” of the firms in the area. Other westerners relied on him for that purpose. That delegation to a trusted holder of tacit knowledge (with presumably a number of such intermediaries existing) can allow a global market to still make use of such knowledge. A real issue it seems to me is Coase’s point about the market vs the firm. Firms are necessarily hierarchical and cannot give their components too much autonomy. And if you act based on orders from above rather than your own initiative, there is less scope for tacit knowledge. Perhaps Scott could write about “Seeing like a C.E.O” or “Seeing like a manager”.

I think language itself is something like a standard, and Hayek was fond of using it as an example. I think even in local situations there are going to be local standards (although Scott suggests they were purposefully malleable in many cases). Some people argue that the extinction of small languages will lead to the loss of some meaning, for languages are not perfectly translateable. But languages are extremely expandable, moreso than most kinds of standards.

One thing I have noticed: Sufficiently fine grained standards don’t interfere with the deployment of local, tacit knowledge.

What about the knowledge of how to repair things: TVs, radios, toasters. Your toaster stops working – you throw it out and buy a new one; standards don’t allow it to be fixed.

More Dogs, Less Crime: Natural language isn’t a standard in the sense used here, because in typical use the speaker and listener are negotiating meaning so there is lots of room for local tacit accommodation. But it is a very interesting example of a very large scale equilibrium that’s not at all like an economic equilibrium! So we should include it in this discussion.

Henri Vieuxtemps: This is an interesting issue but not a negative one for standards. If the parts inside your appliance were standardized you could get replacement parts and local repair people would know how to hook them up. The problem is that each part is customized, the spec and technology change too quickly to track, etc. So actually this example supports an interesting argument that relatively fine grained standards can empower local tacit knowledge.

I think that argument is supported by the creative use of the web, where a set of relatively fine grained standards empowered an explosion of local tacit knowledge.

Henry,

(1) What is the distinction between norms and standards that you’re using? Isn’t a law a norm and a standard? If so, then I’ve already given you an example of a standard produced by the combination of a spontaneous order process and a legislative body, as in Hayek’s Law, Legislation and Liberty.

(2) What’s your evidence of Ostrum’s politics? I just assumed she was something like a classical liberal because libertarians are freaking out waaaayyy more about her Nobel Prize than social democrats. They can’t get enough of her at GMU, for instance. I’m not saying I’m right but that was my evidence, however scant. At least, I had no counterevidence.

It is perhaps worth noting that McDonalds’ menu encompasses numerous variations worldwide.

I see that the CU site is still listing Wilkinson and Lindsey as running things. Is this right?

And as a direct response to Octagon’s query, “How should a classical liberal social theory adapt in order to accommodate this error?”– the theoretical point is the limits of theory. Each standard, each simplification gets produced by a particular historical congruence of actors and institutions with their own goals and limitations, and if we want to know what each misses, we need to drill down into the cases themselves. Sometimes it’s nothing that anyone will care about; sometimes–as in the financial risk models–it’s rather crucial.

To TW: (and sorry I can’t give this reply the full treatment it deserves) – the point isn’t that important things are always lost in standardization. The point is rather that the judgments about what’s important and what can be safely lost are made by particular agents, at a particular (sequence of) moment(s) in time, responding to particular incentives. Others’ interests get incorporated into this, or not–it depends on whether the structure encourages it. In the finance case, both the ecology of how and by whom derivatives were used, and the compensation structures–“I’ll be gone, you’ll be gone”–let to systemically crucial information being destroyed or obscured.

The point isn’t to replace the ‘happy ending’ teleology of markets-efficiently-using-and-creating-information with an equally teleological ‘unhappy ending,’ markets-destroying-information storyline; it’s to emphasize contingency.

“Food standards, whether imposed by government (the US rules restricting knobbly looking vegetables of various sorts) or by large private actors with market power, or by trade associations are an excellent example.”

We’ve had this discussion before, I know. And my complaint has always been (and is always) that I see a huge difference between a standard promoted by, say, a trade association and one written into the law. The example I continually use is the EU rules (it’s a criminal offence to breach these, £5,000 fine and or 6 months in jail) on the allowable components of jam (it’s actually the jam, jellies, marmalades and sweet chestnut puree regulations as revised 2004). One of the definitions (over and above my favourite which is that carrots are defined as fruit) is that marmalades are, by definition, made from citrus fruits.

Now, in a trade body set of definitions we can have “marmalades are normally made of fruits but there are exceptions which require more direct contact between buyer and seller”….as an example. It might be possible to have such in the law but we don’t in fact have that. It is thus illegal to sell pineapple marmalade (I just found out that this is a sometime accoutrement to certain Southern US cooking) in the EU.

You can sell “pineapple boiled with orange juice sauce/spread” but not “pineapple marmalade”. But I would argue that the latter is a description that tells people a lot more about what the product is than the former.

Similarly with the scandium standards I wrote. It’s entirely voluntary to use them: I often don’t myself as I have customers who have much more unforgiving tolerances than could be found in any general standard (I’ve got one who insists on less than 2 ppm Fe and under 1 ppm Zr in his scandium as an example).

Great, fine, we’ve got a general standard which encapsulates certain specific, local, knowledge and anyone can dodge or ignore that any time they like….as long as everyone in the chain agrees that we’re ignoring it.

When the same things are encapsulated in the law such stepping outside the provisions, as a result of having that extra local knowledge that necessitates the stepping out, isn’t possible.

So I would and do argue that industry standards, organically grown standards if you like, ones which grow out of market participants agreeing to have them, are different in kind from those imposed by fiat…whether by the State or by any other holder of sufficient power to insist upon them (and I find it difficult to think of any other actor with that power to insist other than the State……even if Tesco insist that they’ll only sell Grade 1 bananas there are still plenty of other outlets for excessively curved ones….and the Codex Alimentarus where those banana standards comes from only defines what is a Grade I banana. It doesn’t say that other than Grade I are illegal to sell….which the EU implementations do with some of the specs they’ve taken from hte CA).

To go back to Mickey D’s….one regulation which is insisted upon in kitchens is that cooked meat should be stored in refrigerators etc above uncooked meat. This is local knowledge again: uncooked meat tends to have blood in it, which tends to drip. One of the great no nos of any kitchen, commercial or not, is that you don’t allow uncooked foods, especially meats (and poultry even more) into contact with cooked such. And you certainly don’t let the blood from one drip onto the other.

We’ve encapsulated hard won, at the cost of however many upset stomachs and even deaths, local knowledge into a standard.

We’ve also got the more recent insistence that all chopping boards should be made of plastic, not wood. This was a centrally imposed standard, imposed at EU level. Someone, somewhere, thought it was a good idea I guess. It was only after it had been imposed continent wide, restaurants and sandwich shops being closed down or fined for not obeying, that it was pointed out that wooden boards contain bacteria that kill other, infectious bugs (because wood does) and also that washing a wooden board in brine, as butchers have done for generations, cleans wood better than anything at all cleans plastic.

So I’d argue again that we’ve got two entirely different types of regluation/standard. Those which naturally grow out of markets, where they are an encapsulation of local knowledge to the benefit of all and the other sort: the ones that come from clever people being stupid in offices (my definition of bureaucracy).

So when someone says that “hey, standards are all the same, where ever they come from” I’m afraid that I bristle. They’re not all the same: some are hugely useful and some are positively harmful. And my experience is that the useful ones are the market derived ones and the harmful ones are the politically imposed.

“And my experience is that the useful ones are the market derived ones and the harmful ones are the politically imposed.”

Apologies, yes, I know that data is not the plural of anecdote: but then we are talking about local knowledge here. As someone who has both had to implement standards (I spent more years in the restaurant industry than I care to remember) and write them (as above) perhaps that very local knowledge is applicable to the wider point.

As Hayek would have insisted it was of course.

Standardization and uniformization are as likely to result from “market” capitalism as choice and competition. In a recent thread, somebody gave supermarket bread as an example and I argued that the product variety that you find in an average European bakery exceeds that of a Walmart supercenter. George Monbiot, in the somewhat dated article below, points out that the variety of native fruit for sale in a UK supermarket is smaller than it has been for 200 years. This is scary. Biodiversity experts point out that thousands of local varieties of food plants, and their genetic information, have been lost. We do get access to fruit from all over the world that our ancestors never knew but globally, crucial biodiversity is being lost.

Though we still subsist largely on junk, even bilious old gits like me are forced to admit that the quality and variety of most types of food sold in Britain has improved. But one kind has deteriorated. You can buy mangoes, papayas, custard apples, persimmons, pomegranates, mangosteens, lychees, rambutans and god knows what else. But almost all the fruit sold here now seems to taste the same: either rock hard and dry or wet and bland. A mango may be ambrosia in India; it tastes like soggy toilet paper in the UK. And the variety of native fruits on sale is smaller than it has been for 200 years.

George Monbiot, http://www.monbiot.com/archives/2008/09/02/strange-fruit/

Also http://www.monbiot.com/archives/1995/01/01/monotonous-monopoly/, http://www.monbiot.com/archives/2004/10/30/fallen-fruit/

When I was in high-school we were assigned to read After the Fact, which contained a chapter arguing that the push for health & safety standards in food was the result of industry lobbying. I remember we also had to write about an early chapter on colonial Plymouth or Salem with a question (I forget if it was from the book or teacher) referencing “non-political aspects of democracy” which was like discussing a four-sided triangle as far as I was concerned.

This isn’t true either.

“Hayek is remarkably incurious about the actual social processes through which markets work, and in particular the forms of standardization that are necessary to make long-distance trade work.”

You actually haven’t read any Hayek, have you?

Piglet:

Yes, it’s EU.

http://europa.eu/legislation_summaries/consumers/product_labelling_and_packaging/l21134_en.htm

There’s even an amendment dealing with the German Marmelade thing (and with the Portuguese Marmelada as well, which is why you can put apple geranium leaves into qunice jam but not into any other flavour of jam).

Tim, it seems to me there is less constrast between your rare earth standard and the marmelade standard.

You say that pinapple marmelade would need to be sold as for example “pineapple boiled with orange juice”. But would that not be very similar as the results of the rare earth standard? Someone selling something that is not 99% scandium according to the definition of the standard would have to use a more detailed, less standardized description.

You also mention direct contact between seller and buyer as a way to communicate things outside of the standard. But what is stopping people from writing to the Deep South Fruit Company to ask wehre their Pinapple Marmelade is sold, and getting a response back that in the EU it is sold as Pineapple Delight?

The differences seem to come more from different markets more than from the origin of the standards. The far majority of supermarkets shoppers will not read detailed descriptions, nor contact the producer directly. That can make labelling standards overly restrictive to sellers, but on the other hand it produces a pressure to ignore non-binding standards, and put some cheap apples in your 70% Fruit Marmelade

The key point here is that markets use local knowledge only if the costs outweigh the benefits. Veering from established standards often imposes costs, which must be overcome if the venture is to be worthwhile. Hence, “path dependency” arises.

The evolution of a footpath is another example of the same phenomenon. A footpath through a wood may evolve that takes a circuitous route to a destination. But, people may continue to use it because of the extra problems associated with hacking through the wood in a more direct way. (I don’t think Hayek would disagree with any of this).

In the past I’ve worked on designing electronics products for mass production, where I dealt with some of the problems mentioned. Let’s suppose your designing a module to be assembled into a electronic product on a production line. The designer can choose to depend more or less on the skills and knowledge of those working on the line. A product assembly process could be totally standardised within a company and planned so that a low-skilled person could do each task. Or, it could leave more leeway for individual initiative.

There are many choices that can be taken, each with their own risks, costs and possible benefits. A very highly standardised line is complex to design, it requires a great deal of costly process engineering, and it restricts aspects of the product. It risks leaving useful local knowledge on the table. A less standardised line requires more skilled people which is more costly. There is a greater risk of quality problems because of human error.

PCs are a good example of that. The mass-market ones I’ve worked on have used highly-standardised assembly methods because the costs associated with those are amortised over many units. More boutique computers often depend much more on the skills and local knowledge of the assemblers.

Incentives matter a lot here. In a line where people are paid only by the hour there is little incentive for them to veer outside of the standardised process. But, if there is a bonus for productivity then they have more incentive to find quicker assembly methods. But, that type of bonus is a double-edged sword, because it may encourage short-cuts to be taken that harm quality.

At the company I worked for line workers were bonused for productivity. When a new product was introduced there would be an official assembly instruction created by the process engineers. Very quickly the line workers would create there own assembly method. This would contain some genuine improvements. But, it line workers would also find ways to faster assembly that sacrificed quality. (For example, using floppy disk drives as hammers). The process engineering teams would attempt to integrate the improvements into the official instructions and eradicate the bad habits.

Of course there are many mistakes made in deciding these sort of trade-offs. But, how is a centrally planned organization to make those sort of trade-offs at all?

Isn’t it also possible that standards communicate knowledge? I don’t need to invent my own way of integrating HTTP and a public-key infrastructure, because implementing HTTPS guarantees that somebody has already solved those problems.

@ Tim 44
So I would and do argue that industry standards, organically grown standards if you like, ones which grow out of market participants agreeing to have them, are different in kind from those imposed by fiat…whether by the State or by any other holder of sufficient power to insist upon them (and I find it difficult to think of any other actor with that power to insist other than the State……even if Tesco insist that they’ll only sell Grade 1 bananas there are still plenty of other outlets for excessively curved ones….and the Codex Alimentarus where those banana standards comes from only defines what is a Grade I banana. It doesn’t say that other than Grade I are illegal to sell….which the EU implementations do with some of the specs they’ve taken from hte CA).

Seems to me that Apple, Microsoft and the gang have been using market power to impose unhelpful standards for some years now. No-one wants to buy music that only plays on one make of player, or be tied into using one sort of word processor if they don’t want their footnotes to end up all over the page, but for some reason those are the standards we’re stuck with. It’s not an original point, but intellectual property law provides all sorts of ways for companies to impose standards and make it illegal for others to deviate from them.

“Isn’t it also possible that standards communicate knowledge? I don’t need to invent my own way of integrating HTTP and a public-key infrastructure, because implementing HTTPS guarantees that somebody has already solved those problems.”

I think that that’s true. A standard like HTTP doesn’t preclude other people creating non-standard communications systems. But, it provides a standard system, engineers can see if that system is adequate for their purposes and adopt it if it is. The choice of using special local knowledge is always there, if a standard exists that’s just an alternative to it. There may be great costs associated with doing something non-standard, but that simply reflects real costs.

Fuels for combustion engines are a good example of this. An engineer designing a new engine can use one of the standard fuels such as petrol or diesel. In doing that he or she leverages the knowledge associated with those fuels, and the huge distribution networks that exist. Using a non-standard fuel requires much more work, and much more cost. It can only be justified if the benefits of changing are great. That’s why there have been few new fuels introduced for combustion engines.

Or trying to. MP3 is the standard music format in spite of the corporations’ best efforts—they can only nibble around the edges by determining what format to sell in their own stores, but even Apple still has to play MP3s or suffer a catastrophic loss of market share.

Granted: it doesn’t always work. I just thought it was worth pointing out that it’s not like the state has a monopoly on daft standards. Private parties and industry consortia, left to their own devices, can come up with standards that make Brazil look like a Getting Things Done instructional.

If a complete outsider could add a few random thoughts based primarily on study of commodity markets in Egypt in the 19th and early 20th centuries. I think some of you are confusing the idea of standardization with standards, including those for high-quality goods which can be an attempt to segment markets and reap a quality premium. Shade-grown coffee is precisely such an attempt at market differentiation and has little to do with local knowledge as such; it certainly avoids what I believe is a rather crucial piece of local knowledge, namely that children often pick the ecologically sound and tasty beans. That piece of knowledge is hidden for obvious reasons. Large firms (and not markets in a generic sense) often require standardized inputs so they like natural raw materials to be as uniform as possible. I believe they often communicate this to producers by, well, communication rather than markets. In agriculture the uniformity (standardization?) is accomplished by biotechnology (and yes they had a Victorian form of it through plant breeding). Firms pay a premium for standard goods just the way that many people pay a premium for fancy coffees, wines, and cigars. And all of these standards require state regulation to back up the “local” standard–otherwise you get (horror of horrors) Champagne from California, Monte Cristos from Nicaragua, or Brie from Wisconsin. So you have groups of people with specialized knowledge (also called brokers) who supply goods to markets with the required socially produced characteristics. And then you can use financial technology to transform non-standard goods into standard ones–every oil well in the world has a slightly different mix of petroleum but (like wine) it can be blended to a standard (such as WTI or Dubai–which ironically produces almost no oil) and then people pay or are paid depending on whether the actual commodity is “better” or “worse” than the standard. These markets appear to be markets for standardization but also for storage and insurance and generally meet the needs of both large firms and speculators. There are commodities markets in wheat but not tomatoes; in financial instruments but not in automobiles; in chips (silicon) but not chips (potatoes) although there are such markets for raw potatoes (but not I think for raw silicon). These are extremely complex social organizations, much more so (I think) than any other economic institutions studied by sociologists or political scientists. These markets are, essentially, markets in the creation of private money denominated in commodities. I’m told that this is not a possibility because all trades are offset and risk minimized. But I’m not quite so sure given that there have been some dramatic imbalances in such markets on occasion, most recently when the Hunt brothers tried to corner the silver market. For anybody remotely interested I actually wrote a book about this stuff.

Red delicious apples. Standards imposed by growers and buyers eventually resulted in bright red tasteless apples. Most consumers STILL go for it (apples should be bright red they think, and these apples are labeled delicious so they must be), but the more discriminating marginal consumer does not. Thus we have a decade long resurgence of tastier, somewhat obscure, non-standardized, and usually more expensive apples.

I would also note that while commodities may be “standardized” at many points in the supply chain, the end consumer will sort the bundle very efficiently choosing among the standardized product along a variety of predictable dimensions. The bruised and wormy apples well sell last and/or for less regardless of the standard. Seems to me that the market mechanism is at work here as well.