A comment from Gavin Kennedy

Gavin Kennedy writes in the comment thread of this post:
"Daniel

The fact is that Smith's specific (only) reference in Wealth Of Nations referred to some, but not all merchants, from their concerns for the "security" of the capital is sent abroad in the "foreign trade of consumption", preferred to invest in in "domestic industry" (mentioned four times by Smith, twice in the relevant sentence in para 9). This had the "unintended" consequence that it added to domestic "revenue and employment" arithmetically: the whole is the sum of its parts. Smith considered this public benefit. He didn't say anything beyond this. It said similar consequences applied in "many other" situations, without specifying them.
To suggest that this is a general unintended consequence of self-interested actions, leading to "Pareto Optima", "General Equilibrium", as many modern economists do, or that even "selfish" motives lead likewise, is an absurd attribution to Adam Smith. He details again and again how the "self-interested" actions of "merchants and manufacturers" lead to higher prices, less competition, and lower domestic employment in such self-interested" policies as tariffs, protections, prohibitions, monopolists, colonial preferences, the one-sided Combination Acts, the Settlement Acts, Wages set by the magistrate allies of employers, established religions, Primogeniture, Entails, chartered Trading Companies, directly act the general interest, Yet, daily - nay hourly - modern economists are reported, or media sources, continue to pour out nonsense about the existence of an invisible hand in, variously, the market, price systems, supply and demand, and so on.
That lay-people come to believe that in such a fictitious "invisible hand" - let alone that credible figures from our ranks of economists also believe it - though cracks are appearing in the former monolithic consensus sparked of by Paul Samuelson from 1948 - is a disappointing. I look forward to your own recantation of your apparent belied in the fiction of Adam Smith's so-called invisible hand.
Gavin"

Exactly right. I think Gavin is reading me far too strongly. This is the point I was trying to make in my post. You cannot equate Walrasian or Arrow-Debreu equilibrium or any claim about the Pareto optimality of markets with the invisible hand. It is because he equated those things that I was criticizing the linked article. They're not the same thing. It is important to know the general equilibrium properties of competitive markets Arrow-Debreu and all that isn't rubbish. But of course we all know that that's just a model of the real world, it's not the real world itself. And the properties of the Arrow-Debreu system were not what Smith was talking about.

Gavin might have been concerned about my response to bpabbot's comment, where I say that it's a more general point of Smith's than just the specific case of merchants who (to quote Smith) direct their "industry in such a manner as its produce may be of the greatest value". I stand by that. There will be no recantation from me on that. He does generalize this in his discussion of the butcher, the brewer, and the baker, which make the exact same point. Like the merchants directing their industry in the direction that is most beneficial to them, these tradesmen in seeking their own gain do well by others. Smith also cites Mandeville on these points. What is wrong with saying that this is general claim of Smith's?

Look, you can note that the emergence of public good from self-interested action is a general conclusions of Smith's without tying Smith to Pareto optimality or some Panglossian view of markets.

What's most amazing to me is that Gavin could read my post that way, even though I say pretty clearly in the post that: "I don't see how you can read Smith as implying some optimal general equilibrium."

I thought I was being fairly transparent about all this!

Generally Gavin's view of Smith is one that I find more convincing than the way you hear most libertarians talk about him, and this comment of his is no exception.