Saturday, May 15, 2010

Cobb Dug Less Responds

One of the most frustrating things about having a small blog is the searing anonymity of it all. That's why it was so great to see an extensive response to my calculation problem vs. information problem posts on Cobb Dug Less this morning. It looks like Tom is going to respond to each of my posts, but I'll reproduce some of his first post here with thoughts. Tom writes:

"The information in market prices is awesome and they are a wonderful example of spontaneous order. The most important, relevant point is that interference in the markets' ability to freely set prices, unencumbered, has to be the very highest priority."

Agreed. Perhaps I'm guilty of framing it as the "socialist calculation debate," rather than the "socialist calculation debate," but Tom is right here. The point is, even if there were no such thing as socialism Hayek highlights the importance of the price mechanism as something more than one of the two coordinates of an equilibrium (the other being quantity). Price carry information, and even putting all the socialism stuff or even just the "government" stuff aside, that is a valuable insight. I would perhaps raise a question (I haven't entirely thought this through, so I'm not sure what my answer is) about whether this "very highest priority" that commands our attention is a price priority or a relative price priority. In other words, does this priority leave scope for macroeconomic policy? Just food for thought. He goes on:

"I do think that a large portion of the population does believe that government can allocate resources better than individuals. Not economists in general, but this is a widespread belief among a wide swath of the population"

Again, agreed. I think it's safe to say that just about any economist of any persuasion would agree that the general public could benefit from a course in supply and demand, market efficiency, and basic Econ 101 stuff. I'm assuming that anyone reading these posts at least has some understanding of these issues at a level that they don't believe the government can allocate resources better than the market, or if they do believe that they are aware of the oddity of their belief and have some counter-arguments ready (and in restricted circumstances I believe that odd belief as well, and I try to make sure I have counter-arguments ready). That's the assumption that I blog with, but Tom's point about the value of this insight for the general public certainly holds. And just look at my post about the Pew survey or about the recent EJW article... if you have any doubts about the economic illiteracy of the public, those two should dispel those doubts.

"Second, I would argue that while although no one would call Keynes a socialist, he did take an approach that diverged from the classical liberals who dominated economics prior to the Great Depression and placed a heavier emphasis on reliance on government intervention. In some way perhaps, Hayek's essays served to try to influence Keynes to place the priority of market prices higher rather than sacrificing some of their robustness in the pursuit of full employment."

Of course I have to comment on this.

1. Tom is wise enough to recognize Keynes is not a socialist, but the fact that he has to say "no one would call Keynes a socialist" demonstrates that deep down he knows some people do, in fact, call Keynes a socialist! But I've been over this before, Tom is right, and anyone who thinks Keynes is a socialist either doesn't understand (and dare I say, trivializes) socialism, doesn't understand Keynes, or both.

2. I would argue that the "classical liberals" before the Great Depression had different ideas than the classic "classical liberals". I would call Keynes a "classical liberal" in the older sense, albeit certainly with important departures from some of them. And of course he was not a "classical liberal" in the pre-GD sense. But since Tom acknowledged Keynes is not a socialist, I'll let that be for now :) For an understanding of how Keynes saw himself in relation to classical liberalism, I can't recommend enough his book The End of Laissez-Faire.

3. Keynes offered a theory of output that transcended the unsatisfactory "under-consumption theories" of earlier eras. Keynes was not remarkable for his full employment policies. He was remarkable for his theory of output. I think the issue of priorities is a difficult one. Would I have liked to see an economist that had both Hayek's vigorous treatment of prices and knowledge and Keynes's stunning insights into output? Of course I would prefer that. Would I sacrifice the best theory of output we have right now for the sake of further plumbing the depths of knowledge and prices? That is a much tougher question. So we can fault Keynes for this, but by the same token we can fault Hayek for what he neglects. Maybe someone like Ed Phelps is a worthy candidate for the guy that took knowledge, prices, and output seriously.

Generally speaking, though, I feel like Tom comes out in around the same place as me, albeit with different emphases and priorities. His framing of the problem is the same as mine, which gives me faith that I phrased it fairly clearly. Other good reviews from him:

"the framing of the efficient allocation argument has degenerated into a senseless disagreement because of confusion from both sides about calculation and incentives (if I am reading your term incentive problems correctly). Hayek was not writing about incentives in the SCD. The calculation problem exists even in the presence of the so-called benevolent dictator

[...]Unlike Boettke, I differentiate between market failures and information problems. They are overlapping sets, but do not share all elements. Strictly speaking, I think the term market failure is misused as many problems are attributed to market failures, but no market is without distortions. I have no reason to believe that a truly free market won't arbitrage away market failures, but show me a free market
[here I'm a little confused about exactly what he's calling a "market failure"]. Information problems are another story. Information problems is one area in which I split with the Austrians. There is overwhelming evidence that information problems exist, even in the most transparent environments and, as the amount of information has exploded exponentially in the recent past, there is no way to sort through it all. At its most basic level, both buyers and sellers have the incentive to work in their own best interest, but it is much easier to hide information from each other now or mislead and both sides are usually at fault.

[...]Given that the state is fallible, we should not expect a solution, simply a Pareto improvement at best. I would also suggest that any "solution" is an intervention subject to the risk of the Theory of the Second Best. I also invoke Bastiat here: "There is only one difference between a bad economist and a good one: the bad economist confines himself to the visible effect; the good economist takes into account both the effect that can be seen and those effects that must be foreseen."


"What Daniel suggests is directly in line with my thinking about Pareto improvement. However, where we differ is how to deal with it. I am not entirely convinced that a small tax is better than doing nothing. A tax costs overall welfare, a tax on carbon emitters has enforcement costs, and a broadly applied tax has free rider problems. I would be in favor of tax cuts for carbon emitters and shift the burden of proof of reduction of carbon emissions to the companies if they want to take advantage of the tax cuts, vastly reducing enforcement costs."

This is all great stuff. I'm not involved in the environmental economics literature, so I don't know the best way to implement this. This tax cut idea is completely fine - I'm not picky about the implementation. My point is that there is a case for a correction of some sort.

I would push back a little on Tom's statement that "I am not entirely convinced that a small tax is better than doing nothing". I'll grant all his enforcement cost and dead weight loss points, but its important to recognize that right now, with no correction to this incentive problem, our liberty to breath clean air and live on a healthy planet is being imposed upon. The lack of property rights not only causes this problem - it also removes our ability to make a broadly recognized claim to a violation of liberty. It seems silly to say "my liberty to breath clean air is violated" because our conceptualization of liberty is so solidly grounded in property (for good reason, let me qualify). But let's not take that conceptualization too far. If we rely too heavily on that, liberty becomes a question of the luck of the institutional draw, and it loses its fundamental meaning and importance.


  1. I have been thinking about knowledge problems vs. incentive problems a little more. Some time ago I thought about this distinction in more depth, but I don't write things down, move on, and forget about them.

    The issue seems to be the sensitivity of decision-makers to feedback.

    For knowledge problems the decision-makers are insensitive to feedback. The picture they receive of the world is lagged and low resolution, or at least too much to solve complex coordination problems. Decision-makers may be sensitive to the "right" kind of feedback, but they just can't get enough of it in high enough quality. By decentralising decision-making we can "get the whole picture," though only the eyes behind the invisible hand see it all at once.

    For incentive problems, decision-makers may get very timely and high resolution feedback. The problem is rather that they are sensitive to the "wrong" kind of feedback. Some economically important fact is not not being registered as relevent to decision-making, i.e. with our imperfect property rights regime, something is being left out of the calculation.

    Another way of putting it would be that if agents see the whole picture, but it is too lagged and low resolution, then we have a knowledge problem. Alternatively, if agents don't get the whole picture--even if what they do get is immediate and high resolution--then we have an incentive problem.

    In either case, the problems boils down to decision-makers not being sensitive to feedback, but in each case for a slightly different reason. I think this is slightly clearer than I wrote the other day.

  2. That makes a lot of sense.

    Here's one thought, though - again taking the familiar (albeit admittedly easy to think about) - example of carbon consumption.

    When I think about this example it really makes me veer towards siding with Hoppe and Mises over Hayek. Lot's of people know about the concept of a negative externality and how it applies to carbon consumption. Even people who don't know what an externality is have a sense of knowledge that we are consuming a sub-optimal level of carbon. The knowledge isn't the problem at all. What's missing is an incentive to change behavior - the property rights - the stake in the issue. It wouldn't surprise me at all if the most informed climatologist in the world drove a gas-guzzling car, precisely because this doesn't seem to me to be a problem of a lack of knowledge so much as a lack of incentive to act on that knowledge.

    That's not to say your schema doesn't fit a lot of problems. In many ways it's relevant to things like asymmetric information problems. But I think there are some important cases where knowing what the problem is isn't the concern - lots of people know and still don't act on it because they have no stake in it.


All anonymous comments will be deleted. Consistent pseudonyms are fine.