September 2022

Sun Mon Tue Wed Thu Fri Sat
        1 2 3
4 5 6 7 8 9 10
11 12 13 14 15 16 17
18 19 20 21 22 23 24
25 26 27 28 29 30  
Blog powered by Typepad

« Another Brick in the Regime Uncertainty Wall | Main | A Tribute to Teachers Outside of the Classroom »

Comments

Feed You can follow this conversation by subscribing to the comment feed for this post.

The concern is he makes inappropriate inferences about macroeconomic adjustments from relative market adjustments. Yes - we continue to employ more youth relative to old people in the summer and we continue to employ more youth in the summer than at other times. But this relative demand for youth doesn't tell us anything about total demand for labor. This is what most of Mulligan's posts seem to consist of - telling a pretty good relative demand story and then in the conclusion forgetting he's told a relative demand story.

What's amazing to me is how transparent it is most of the time, which is I think why people keep coming back to his awful summer employment post. You see the blips during the summer in employment but you also see that every single summer has lower total youth employment than the previous summer. His own graph blares out "relative prices are different from aggregate prices - relative demand is different from total demand", but it seems to be completely lost on him, which is amazing because he's a very talented economist.

This is my most recent reaction to Mulligan's most recent teen employment post: http://factsandotherstubbornthings.blogspot.com/2011/06/casey-mulligan-on-seasonal-employment.html

Daniel,

You are not getting the point. Total demand and aggregate prices are not economic phenomena that affect human action, other than by policy makers. They are statistical constructs that don't have real economic counterparts. Just because I can measure the total height of all people in the world doesn't mean it has any meaning.

J Oxman -
Of course they have real economic counterparts. The general price level is just the reciprocal of the value of money, which is definitely a real phenomenon. And total demand is just all the micro demands added up. These are absolutely real and crucially important phenomena.

And to the extent that micro has implications for them, we need to pursue micro answers.

Mulligan's problem is that he takes a half-filled balloon, squeezes one end till there's no air in it, watches the other end fill up with air, and declare there's no balloon inflation problem because he observes a relative differential in the air at the two ends of the balloon.

That's not good economics. I agree with Peter that we need good micro to work wtih good macro. My problem with Mulligan isn't that he tries to provide micro insights - it's the quality of the micro insights that he provides.

Daniel,

Value of money is terms of what? Snickers bars? Ford automobiles? Help me out here.

My point is you cannot add up the micro demands and get to a meaningful amount of something. It might look meaningful, because we can do it in terms of dollars and cents, but it misses all the important differences among these various demands.

Thinking in macro terms papers over the important differences, like relative prices, that affect real behavior, like the decision to hire some construction workers to build new houses.

I don't follow Mulligan's blog, but his research papers are excellent. I find your statement that he offers low-quality insights highly suspect. I think it much more likely that you disagree with his insights, but you are arguing on different terms.

If aggregate economic data are not economic phenomena, then why is everyone, you included, trying to explain it? For something that has no meaning, there is an awful lot of words being spilled over it.

re: "Thinking in macro terms papers over the important differences, like relative prices, that affect real behavior, like the decision to hire some construction workers to build new houses."

So people always say this and yet I deal with micro issue every single day... maybe I'm innoculated... maybe it's something else. But I don't see why macro supposedly makes micro considerations just vanish. Nobody criticizes Mulligan because he talks about micro stuff. They criticize him because he talks about it badly.

re: "I don't follow Mulligan's blog, but his research papers are excellent. I find your statement that he offers low-quality insights highly suspect. I think it much more likely that you disagree with his insights, but you are arguing on different terms."

Right - this is part of the reason why I said his blog posts are so strange because they seem like they're written by a different person. I think you should read the blog posts and some people's responses before making assumptions about me. As Peter suggests - I'm hardly alone in scratching my head over the Mulligan posts.

J Oxman -
This point, "Thinking in macro terms papers over the important differences, like relative prices, that affect real behavior, like the decision to hire some construction workers to build new houses" seems like an especially problematic one for you to make since all of Krugman's answers in the link that Boettke provides are also micro responses!!!!

Daniel,

I don't read Krugman.

Martin,

As Roger Garrison said, there are macro questions but only micro answers.

Furthermore, if someone says "unemployment" is up, my first question is "unemployment of who?"

In other words, I'm not trying to explain macro phenomena. I am trying to explain real economic phenomena. Statistics can sometimes be reasonably good proxies for something, but they are not that thing in and of itself.

I'll have to read Mulligan's blog myself, but I have my own suspicions about macroeconomics. I think there is a macroeconomic reality, but almost nobody has ever actually written about it, excepting those who have written about the emergent properties of spontaneous orders. Everything considered "mainstream macro" is mostly a bunch of mysticism without much meaning -- there is a long history of gallons of ink being spilled on nonsense, so that's no argument for mainstream macro. Much ink was spilled over aether, alchemy, and Ptomaic astronomy as well.

J Oxman, that's great, but you seem to be arguing that there are no micro answers for the movement of aggregates, yet you come out and defend regime uncertainty as the explanation for said movements.

"My point is you cannot add up the micro demands and get to a meaningful amount of something. It might look meaningful, because we can do it in terms of dollars and cents, but it misses all the important differences among these various demands."

If adding up the micro-demands is not meaningful, then what is it that you're doing?

On a side note: http://en.wikipedia.org/wiki/Sonnenschein%E2%80%93Mantel%E2%80%93Debreu_theorem

Troy - there are lots of systems characterized by emergence where the emergent processes themselves aren't always the primary subject of research. I don't think this makes them "mystical". Krugman himself has written a lot about emergent order, after all. He's the one that introduced me to the idea, and I only came across Hayek afterwards. Clearly he doesn't think there are any great contradictions. I'm entirely accepting of the macroeconomy as an emergent process and I've never noticed any contradictions between that and a lot of mainstream macro. I wish I knew exactly what you thought the contradictions were, because you never seem to be able to get past the initial accusation. Too much trendy, under-developed counter-establishmentism manages to pass itself off as "emergent order", I think. What exactly are you thinking of?

Interesting that Krugman admits that there's "more to the Texas story ... like oil" -- but not that there's "more to the Texas story -- like no state income tax."

I side with Daniel. Further, Peter asks the following: "And, if I am not missing anything, then how can someone build a "general theory of anti-Mulliganism" without rejecting the use of microeconomics to understand macroeconomic problems?"

What about this: http://noahpinionblog.blogspot.com/2011/08/possible-resolution-of-krugmanmulligan.html

Are Austrians inclined to accept this?

Mulligan is the guy who explained the sharp rise in unemployment in late 2008 and early 2009 as being due to an exogenous shock to the intertemporal time preference rate of workers. A whole bunch of them just got lazy all of a sudden and stopped working, apparently because they foresaw that Obama would get elected and raise their taxes, although in fact he cut taxes, although actually expecting a rise in taxes would say that you should work really hard now while the taxes are low and quit later when they actually go up. That all the stats show an increase in the number of workers out of work who were seeking work, well those are probably phoney macro stats since all that matters is micro, and those ones who quit, well, they did not foresee that they would run out of money and want to get a job pretty quickly again after quitting, just a bunch of lazy dumbies whom we should not worry about anyway.

Yeah, Mulligan, such an insightful economist...

Krugman isn't a macroeconomist -- he's a spatial economist -- and it is quite unfortunate that 1) he thinks that because he's good in his specialty, that means he fully understands all of economics equally well (see Socrates' points on why that is unwise in Plato's "Apology"), and 2) he doesn't in fact apply the insights of self-organization to macroeconomics. The weakest part of "The Self-Organizating Economy" is the part on macroeconomics, where he abandons everything he had just said, and turns to Keynsianism.

Anyone who believes that top-down control of a system that one is an internal part of engages in mysticism. That doesn't mean that the emergent qualities of the system in question is mystical. Positing that living things are products of evolution and self-organization is science; positing that they are products of intelligent design or special creation is mysticism. Belief that you can transform iron into gold is mysticism, and so is the belief that one can transform an recession into prosperity through things like printing money, deficit spending, digging holes and filling them, building armies to fight Arabs or aliens, bailouts, or continuing to do the same things that cause the recession in the first place. And believing that you can understand a system from the outside when you are inside that system is either a demonstration of mysticism or ignorance -- but one cannot argue that it is the basis of a science. The best one can do is extrapolate what seems to be happening through the creation of analogies with similar systems of less complexity than ourselves. At best. But anyone who says "the macroeconomy has this or that feature" is not a scientist, because they are engaging in mere speculation about something they inherently cannot understand, being more complex than them. You might as well be theorizing about the flying spaghetti monster.

I have been critical of Krugman on numerous grounds, but actually his original field was foreign exchange rates, that is to say, open economy macroeconomics. There has long been a body of opinion that he deserved his prize more for that work than what he did get it for, given how much of that was done by others earlier, hack, cough. So then he went into micro trade theory and after that into the "new economic geography," or "spatial economics," if you prefer.

Troy -
Certainly anyone who believes in top down control of a system doesn't seem to understand principles of emergent order, but that doesn't answer my question - what's the contradiction between mainstream macro and emergent order.

re: " Belief that you can transform iron into gold is mysticism, and so is the belief that one can transform an recession into prosperity through things like printing money, deficit spending"

Again Troy - can you please explain WHY you think this rather than simply repeating the claim?

The conflict is that emergent orders have emergent properties. You cannot predict the behaviors of a cell from a mere understanding of organic chemistry. Organic molecules, which are incapable of having goals, give rise to cells which are capable of having goals. That's a pretty significant difference. We can say that a cell can "hunt for food," but we cannot say that a molecule can "hunt for a ligand." Molecules just bump into each other and, if the reactive parts are close to each other, react. Enzymes make this process more likely, but enzymes are a product of cells -- which are in fact capable of hunting down ligands, aka food.

Now imagine that you are a conscious amino acids, which means you have no concept of goal-oriented behavior. Yet you are trying to make sense of your surroundings. You can have a "microchemical" theory of how reactions and interactions take place, because those are clear and known and obvious. However, even though there is a "macrochemical environment" (the cell), the best you could do is get a sense that there is something going on. Local patterns -- such as organelles -- can be made out, but there is no way to get "outside" the cell to understand it in its totality. From the POV of the amino acid, any "macrochemical theory" would be mostly speculation. One might speculate that the larger macrochemical environment looks a lot like the mitochondria, but even then the analogy would be only partial. No complete theory would be possible.

Further, let us suppose that the amino acid comes to realize that zinc is an important atom for the macrochemical environment. He sees it being used in gene regulatory proteins and a variety of other enzymes. When zinc is low, some proteins are not made, and that affects the creation of other proteins, etc. A slowdown in protein production would be similar to a recession. The protein figures this out, and is somehow able to bring much more zinc in. At certain concentrations, zinc is a poison, though -- something the amino acid cannot know, not understanding the true nature of the macrochemical environment. Thus, the health of the cell is made worse.

Now, if you have problems understanding how a goalless amino acid thinks, then you have problems understanding how a goalless spontaneous order works. That is what we are faced with: a goalless spontaneous order created by goal-seeking agents. We are on the interior of that system, and cannot understand what it looks like from the outside. It is impossible. To the extent that mainstream macro claims otherwise, it is wrong in the most fundamental way. One can look at smaller, similar systems -- like cities -- and draw conclusions regarding the similarities between cities and the overall economy, but one still will not have a real full understanding of the macroeconomy. To believe that one can positively affect the macroeconomy through pushing that button or pulling this lever is to believe that top-down control is possible. The system is not even cybernetic, because such a system had negative feedback only, and thus can be steered. The system is a combination of positive and negative feedback simultaneously, and thus cannot be either controlled or steered. The tiniest of differences between one time and another makes it impossible to know if what you want to do will work or not, or to what degree it will work, or what unforeseen consequences will result. So when someone says that printing money or deficit spending will have this or that effect, they are blowing smoke, because it is inherently impossible to know that such will occur in every situation, because every situation is in fact different in the macroeconomy. This is not to say that microeconomics is similarly impossible. Humans do act in pretty predictable ways when we know the circumstances. However, while each person can know his own circumstances pretty well, nobody can know the circumstances of everyone, and that is what is required to do macroeconomics. Which amounts to mysticism.

Now, I can sit here and explain why printing money and deficit spending won't help an economy, but I cannot do it using the bogus mathematical formulas used in macro (one can use math to "prove" Ptolmaic astronomy too, but that doesn't make it true -- got epicycles?). What I can do is point out that the money will be directed not to where it needs to go, in no small part due to the fact that nobody can know where that is anyway, but mostly because of the problems of public choice. Money will be directed toward political cronies and/or big, visible corporations (GM and Chrysler come to mind). But these places are not where economic growth comes from. Economic growth comes from innovation, and innovation occurs not in large companies, which do a good job of making known things, and do so increasingly cheaper, mostly through increased efficiencies that reduce the numbers of employees they need. No, innovation occurs in small companies, which hire large numbers of people. Government money is never directed into these places. Thus, government money cannot contribute to economic growth. All government money does is support already-established companies, which may maintain the economy where it is at, but do not contribute to much if any real economic growth. Thus our stable unemployment rate at around 10%.

Troy - I understand the concept of emergent order fine. But what you've presented here seems to me to be a reasonable argument against (or at least an argument for skepticism about) microfounded macroeconomics. Building up from the micro level in complex systems is very tenuous business.

You apparently think that people are arguing that we can control or know in advance how individuals will react to monetary or fiscal stimulus. You clearly have a different view of macroeconomics in mind than I do if this is what you're arguing against.

I am arguing that the best we can do is microfounded macro, since we can only in fact know about the microfoundations and cannot know anything about the emergent features of the macroeconomy. I am arguing that macroeconomics, insofar as it attempts to deal with the emergent properties of the economy, cannot be science, because we are not more complex than the economy, but less, meaning we cannot understand it. To do so, one would need a God's-eye view of it, and that is impossible.

Now, if you agree that we cannot know how individuals will react to monetary or fiscal stimulus, why recommend it? Such recommendations do in fact come out of macroeconomic theory, and I am arguing that one cannot have an accurate macroeconomic theory precisely because one cannot take the God's-eye view necessary to have such a theory.

Take the amino acid view again. How could you know if you are in an ameoba, a sponge, or a human? You can't -- but it really makes a difference which one it is for the purposes of nutrition. You do not and cannot know anything about the macroeconomy that would result in being able to make any suggestion as to what the government should do to direct or steer it. If you don't understand that, you don't actually understand the concept of emergence.

"Just because I can measure the total height of all people in the world doesn't mean it has any meaning."

Uh, curious, because you just said what it means!

"Thinking in macro terms papers over the important differences, like relative prices, that affect real behavior..."

This is about the scientific equivalent of declaring that we can never discuss thunderstorms, because the only "real" thing is the movement of air molecules.

"I am arguing that macroeconomics, insofar as it attempts to deal with the emergent properties of the economy, cannot be science, because we are not more complex than the economy, but less, meaning we cannot understand it."

So history is impossible, and all of the departments of history in the world should just go home. (Obviously, history, being made up of many individuals, is more complex than any one of them!)

It is amazing what utter nonsense people can convince themselves of if it supports their ideology.

Well said Gene.

Methodological individualism ought to discipline us in not ascribing agency to aggregates, but it's always sad to see someone distort that important insight into an unscientific allergy to aggregates and the careful study of their behavior and the emergent behavior of their constituent parts.

Gene,

How can you accuse someone of ideological fraud while making such a non-sequitur? History is philosophy played out! We don't have to wonder, guess, calculate, predict, etc. what would happen if Ceasar crossed the Rubicon because he DID crossed it and we know what happened afterwards so history has nothing to do with what Troy explained: "it is inherently impossible to know that such WILL occur IN EVERY SITUATION, because every situation is in fact DIFFERENT in the macroeconomy."

Daniel,

I usually enjoy your comments on this blog since I consider them to be thoughtful and well-reasoned. You asked of Troy: "can you please explain WHY you think this rather than simply repeating the claim?" Troy complied so now I asked of you to explain what is wrong in Troy's story or at least how is your understanding of macroeconomics different from Troy's.

Gene,

History is not the story of aggregates, it is the story of individuals acting in specific situations, especially those people whose actions had an effect on large numbers of other people (much postmodern historians' work notwithstanding). Thus, history is not impossible, as it is the study of those actions. However, if you consider the study of history to be an attempt to discover the "patterns of history" ala Marx or other "scientific" historians, then you are in fact not doing history, as such patterns cannot be discerned by us, being, again, inside the system. If such patterns exist, they can only be discerned by a mind more complex than the social-historical system of which we are a part, but not by us. This isn't ideologically driven -- it is the science of complexity.

One can study aggregates that one is external to -- such as the aggregate of cells that make up an organism -- but one cannot study such aggregates when one is internal to the system and understand the emergent behavior of that system, as I demonstrated in the above examples.

Gene,

I can define something but when I say it has no meaning, I intend that the object is not an important bit of information for human action. Total height of all humans is a definable term, sure. But it's completely irrelevant to anything.

As for the comment about macro papering over the relevant things... a thunderstorm is a real thing. I can add up all the cotton grown in the U.S. - now that's a meaningful aggregate. But total dollar spending? That's not a meaningful aggregate. There's so much action within "total spending" that to just look at falling GDP and say "we need more spending!" is, to me, nonsense.

I'm not, as DK would say, "unscientifically allergic" to aggregates. My concern is that it is the aggregates that are the focus, not the real action that is generating those aggregates. If I state the case too vociferously against aggregates it is because they are too often the stopping point.

I place more confidence in microeconomics and find it easier to understand (there's probably a connection!). I actually like a lot of Mulligan's posts because he's saying the normal laws of economics continue to apply. But at the same time the "quasi-monetarists" have persuaded me that a change in the nominal money supply has real effects (though this phenomena must ultimately have micro causes) and so one needs to grapple with macro when addressing macro questions. Nick Rowe is a good explainer and comments on the recent Mulligan dispute here:
http://worthwhile.typepad.com/worthwhile_canadian_initi/2011/08/aggregate-supply-and-relative-supply-and-demand.html

Eric -
I did respond, didn't I? See my August 16th comment.

Troy makes some very strange arguments about the implications of economic complexity. It would seem to me the real insight we can derive from it is a bias against microfounded macro - or at least a skepticism of it. He seems to come to just the opposite conclusion.

He likes to go down to the cellular level (or even sub-cellular level), and that's a great way to think about this. Organisms are complex systems formed by combinations of cells. Ecosystems, in turn, are complex systems formed by combinations of organisms. Ecologists seem to get along fine without building up their understanding of ecosystems from micro-foundations based in biology. Biologists who study organisms (as opposed to cellular biologists) seem to get along fine studying organisms without building up every insight from the cellular level. Indeed - BECAUSE these are complex systems one would expect two things:

1. The systems will probably exhibit emergent behavior, and we should be aware of that and because of this

2. It's extremely hard to build up an understanding of the system as a whole from microfoundations.

And yet it's PRECISELY #2 that Troy repeatedly advocates on here! I've made this case above, Eric. I'm responsive to Troy. Troy is too eager to accuse others of being mystics or Ptolemaic (his all time favorite insult) when he doesn't seem to understand the implications of economic complexity himself.

And notice I'm not going around calling Troy an adherent to mysticism or Ptolemaic or anything like that. I'm simply arguing he's in error. I don't know why Troy always feels the need to go that route.

TGGP -
I think that's exactly the right way to think about it, and Nick's post was very good.

Nobody is arguing that Mulligan's points were wrong, you realize. The criticism is that he's drawing inappropriate conclusions from his observations about relative prices.

No question money supply has real effect, much like the quantity of any given good has an effect on prices. And so does demand. But these effects are more local than we like to admit when we think in macro terms. Further, we end up with central planning of the money supply for large numbers of people over a wide area, when what we really need is decentralization of the money supply, responding to local demand, which differs from place to place. The way we do it now creates instability, because it may be too much for one place, not enough for another, and just right for another, with inflationary and recessionary results in different places and different sectors in those places. There is also the problem of the money going not to where it is needed, but to where politicians want it (first stop of much created money: U.S. Congress). If money and most other elements that are part of macro were decentralized, it would be pretty much nothing but microeconomics everywhere, and the study of spontaneous orders would be "macroeconomics" (but really, sociology).

Troy -

re: "The way we do it now creates instability, because it may be too much for one place, not enough for another, and just right for another, with inflationary and recessionary results in different places and different sectors in those places."

See but you go around blustering and bad mouthing mainstream macroeconomists and then say things like this, apparently not realizing that there's a massive literature on this phenomenon compiled by mainstream macroeconomists, and discussed in the recent past by people like Mankiw and Krugman.

You make these criticisms that show little awareness of what mainstream economists actually work on and then you dismiss them as being mystics or Ptolemaists when either (1.) everyone is well aware of the correct points you make, or (2.) your points are wrong.

You always remind me of this: http://xkcd.com/675/

By all means, point out where I am wrong. Give me specifics, so I can respond specifically. I am arguing that the only things of any value in macro can be understood using micro, that most of what is exclusive to macro is nonsense, and that the only macro theory is really a sociological theory. Now, you can disagree with all of that, which is fine, but 1) that doesn't make it wrong, because not everything you believe is necessarily right, and 2) I have learned this stuff a bit longer ago than an hour. You managed to find an economics theory that supported the folk economics you were born believing, and think it's science. Math doesn't make something a science. Ptolmaic astronomy had quite a bit of math supporting it. It was a very predictable theory -- at least for a while. Keynesianism reminds me of it precisely for that reason. It seems to make sense, it's not entirely counterintuitive (any idiot can see the sun goes around the earth; any idiot can see that it's because people just won't spend their money, for some strange reason), and over the short term it seems to be correct and give correct predictions. But in the end, it is just deeply, fundamentally wrong.

This is probably not a tangent I should go on, because you're just going to mangle it, but Ptolemaic astronomy WAS good science, it just didn't turn out to be right. I find it interesting that people always bad-mouth Ptolemaic astronomy or phlogistan chemistry or whatever else. These were good, progressive, scientific attempts at explanation that were displaced by better, progressive scientific attempts at explanation. Dismissing them like they were idiots is naive because in all likelihood we're going to have other paradigms in the future that do an even better job explaining our world than our current paradigms. That fact doesn't make what we're currently doing unscientific. Indeed - the science we're currently doing is crucial to eventually ushering in future, better paradigms.

re: "You managed to find an economics theory that supported the folk economics you were born believing, and think it's science."

Don't pretend to know my thought process Troy. I don't adhere to folk economics, I certainly didn't come to economics with Keynesian intuitions (quite the opposite). There is no point to me adopting scientific views simply because they reinforce what I want to think.

re: "Math doesn't make something a science."

Quite right. Has anybody been making that argument Troy? Why do you bring this up?

That's your interpretation. I'm not describing them as idiots. That's you trying to impose that on me, and you cannot say you got it from anything I have said. We now know it is wrong. When a better, more descriptive theory came along, it was replaced. A more accurate science replaced what was a less accurate science -- based on the facts at hand. Those who remained Ptolemaics after Copernicus, Kepler, and Galileo were doing so for reasons other than scientific (which can still be dangerous, as Galileo learned).

Keynes is mathematized folk economics. That is why people believed in it before Keynes, and why Keynes, in giving it the veneer of respectability with math, became so popular. Politicians were Keynesians before Keynes, and it's not because they were good economists. Keynes was so popular precisely because he seemed to confirm people's preconceived notions about how the economy really worked. My point in comparing it to Ptolemaic astronomy is that with Ptolemy you had someone who was able to explain mathematically what appeared obvious to any observer on earth: that the sun went around the earth. Folk astronomy was thus made scientific through math. But the fact that math can "prove" Ptolemaic astronomy doesn't mean it is in fact true. It turns out that folk astronomy is wrong -- and so is folk economics. And no amount of math makes it true.

Troy - with all this talk about the necessity of belt-tightening and living beyond our means and wasteful government spending, I don't think Keynesianism is the one guilty of ratifying a folk economics.

And could you stop saying the fact that Keynesians use math doesn't make it scientific? This is obvious Troy. Nobody ever said that's what makes it scientific.

That's the only thing I can find that could make anyone think it's scientific. It's certainly not because it's actually science.

Folk economics argues that spending is what is most important in the economy, no matter what the spending. There's no real understanding that creating new stuff is the real driver of economic growth. Heck, that's considered to be an "external shock". Nonsense.

"Folk economics argues that spending is what is most important in the economy, no matter what the spending. There's no real understanding that creating new stuff is the real driver of economic growth."

First of all that is NOT folk economics -you could argue that it's a crank economics that has captured academia (I have no idea) -you could argue that it's a propaganda that serves particular interest groups at particular times (I think that's a fair point), but you can't really call it folk economics.

"The government should make a budget like my nuclear family does, and stick to it" is folk economics. "The rich are stealing from us and that's why we don't have more money" and "The lazy poor are parasites on us, and that's why we don't have more money" are also folk economics.

I don't think Keynsians would argue that it makes no difference if stimulus money goes to infrastructure upgrades or casino subsidization, but correct me if I'm wrong.

From what I've read on folk economics, it most certainly is folk economics.

And the Keynesians in fact do argue that it doesn't matter where the stimulus money goes, because it will end up throughout the economy eventually. That's what matters, not necessarily the long-term value added of infrastructure (though this is only for getting one out of the recession -- they are more sensible during regular times, when they would in fact argue for long-term value added).

The comments to this entry are closed.

Our Books