Tuesday, April 14, 2015

Did macro theory fail us in the crisis?



David Andolfatto and Mark Thoma have posts defending macro theory from (some of) the people who say it failed us in the crisis. Both are good posts, and you should read both.

Anyway, here's a quick (and probably incomplete) taxonomy of criticisms people make about macro with regards to the crisis:

1. "Macro models failed to predict the crisis, therefore DSGE sucks."

This is the criticism that Andolfatto and Thoma reject. I basically agree. There are no other models out there that did forecast the crisis. Nor are expert predictions any better.

Personally I think DSGE techniques haven't  reaped dramatic benefits (yet). But what other alternative is better? When I ask angry "heterodox" people "what better alternative models are there?", they usually either mention some models but fail to provide links and then quickly change the subject, or they link me to reports that are basically just chartblogging. Yeah, sure, if you put out hand-wavey reports saying "capitalism sux, there's gonna be a crash!" every year or two, you're eventually going to be able to say "see, I told you so". But that's no replacement for real modeling.


2. "Macroeconomists were too confident before the crisis, and that gave policymakers false confidence."

It is pretty obvious to anyone who has ever interacted with macroeconomists that most of them take their models way too seriously (this is even more true of the "heterodox", to the degree that they even have models). It's a common disease of academics in general - you have to spend so much effort pushing your theories that overconfidence is selected for.

Did this confidence leak over into the policymaking sphere? I don't have evidence here, but I doubt it. Most of the Fed people are a LOT less confident than academics. And they were being advised by a lot more than just the academic crowd - they had a big stable of chartbloggers, hand-wavers, etc. to draw upon. Plus they themselves had Old Keynesian models in their bag of tricks. As for politicians, it's not clear they even know that academic macroeconomics exists.

If the Fed people were overconfident in 2005-6, I suspect it was mostly due to natural cognitive biases - "Everything seems like it's been going OK for a couple decades, I guess we're doing something right" - rather than the overconfidence of the academics they interacted with.

But I could be wrong.


3. "Macroeconomists weren't focusing on finance enough before the crisis."

Thoma says that this is a valid criticism, and I agree. There are a bazillion models out there. But just having models out there isn't enough; if you're going to give policymakers real advice, you're going to have to choose which model - or which basket of models - to base your advice on. Macroeconomists weren't very worried about finance before the crisis - you didn't see a lot of people waving copies of Geanakoplos (2003) and saying we could be courting disaster.

Belief in the Great Moderation, and in the Fed's ability to stabilize the economy, was too strong. The central problem of depression prevention had not, in fact, been solved. But an awful lot of top macroeconomists - not just Lucas, but the New Keynesians too - thought it had. Their favorite models didn't have any finance in them, with the possible lone exception of the Bernanke-Gertler "financial accelerator" models.

That was a big mistake, especially since the Great Depression and crises in other countries (e.g. Japan) should have suggested that financial crashes were a big deal. To their credit, though, mainstream macroeconomists have been hastening to correct the mistake. Certainly they're going to pay more attention to finance for at least a few more decades.


4. "Macroeconomists don't do enough to kill their models off."

This is something I hear surprisingly few people say, given that I think it's the best of the criticisms out there. If you let a million flowers bloom but don't cut any of the flowers, you get a big warehouse full of flowers. OK, so that metaphor went nowhere, but you get the point. Macroeconomists, when they get defensive, tend to say something along the lines of "We got models for everythin'!" But is that a good thing??

I feel like if you have models for everything, you don't actually have any models at all. Without a way of choosing between models, your near-infinite stable of models turns into one big giant mega-model that can give anyone any results he wants. Worried about a financial crisis? Pull out a model that tells you a financial crisis could be looming. Worried about inflation? Pull out a model where inflation is a big danger. And so on.

Now, technically, you could choose between models based on the plausibility of the assumptions. But three things make this impossible in practice. First, the need for tractability means that the assumptions in almost any modern macro model will be utterly implausible to anyone who has not spent decades in a monastery high in the Himalayas training himself in the art of self-deception. Second, the assumptions are so stylized that it takes a huge amount of talent just to figure out what they are - in fact, we're starting to see the emergence of top macro people, like Matt Rognlie, who specialize in figuring out what the heck models are actually saying. And third, with a near-infinite catalog of models to comb through, there's just no way to compare any significant number of them all at once.

If you ever want macro models to actually be useful, it's not enough to just wave your hands and say "all models are wrong". It's not enough to treat models as ways to "organize our thinking". You've got to have a way to take them to data and decide if you should keep them around, send them back to the shop for alterations, or burn them in a fire.


5. "The crisis exposed the fact that macroeconomics doesn't work."

Well, sure. But it also showed that we need to keep trying to make it work. And macroeconomists, as a whole, don't absorb a significant fraction of our GDP, so I'm not incredibly worried.



Updates

Bumped from the comments, by an anonymous commenter:
[D]on't focus on macro *theory*. It's macro empirics I'd worry about. theory ahead of measurement etc. The difference post crisis is you see greater prominence of macro papers using micro data (e.g. mian sufi or autor/dorn/hanson/acemoglu).
Great point.

105 comments:

  1. "Belief in the Great Moderation, and in the Fed's ability to stabilize the economy, was too strong."
    -No. As Scott Sumner has frequently pointed out, it was too weak.
    "The central problem of depression prevention had not, in fact, been solved."
    -Agreed.
    "If you let a million flowers bloom but don't cut any of the flowers, you get a big warehouse full of flowers."
    -Or a big warehouse full of weeds.

    ReplyDelete
  2. Here's my model that predicts the crash. Maybe not this one, or the next. But the one that will matter.

    http://www.amazon.com/Capital-Critique-Political-Economy-Classics/dp/0140445684/ref=sr_1_1?ie=UTF8&qid=1429046013&sr=8-1&keywords=capital+vol+1

    ReplyDelete
    Replies
    1. Like the one in Cuba in the 1990s?

      Delete
    2. Anonymous11:48 AM

      That is not a model, it is gibberish.

      Delete
  3. Maybe I'm one of the surprisingly few. My beef has been that economists in general, micro as well as macro, don't take Type I error seriously enough. If you're an economist you can say all kinds of stuff, and if you're wrong, you can just go on saying the same stuff and no one gets upset.

    If you look at the results in Fanelli's Hierarchy of Science piece from 2010, economists actually rank worse than most other social scientists, and well below most natural scientists, in publishing likely false positives. (This is an assigned reading for the stats class I'm teaching tonight, so it's fresh in my mind.)

    ReplyDelete
    Replies
    1. I think you're right.

      Delete
    2. Matt Molewski1:54 AM

      That's more or less been the gist of virtually every blog post Paul Krugman has made lately, regarding the many who predicted skyrocketing inflation due to Fed policy during the recession, and who apparently haven't been confronted with the fact that they were absolutely wrong. I can't speak for the profession as a layman, but it seems that popular opinion on economics tends to gravitate towards whatever view most suits a given bias: That's a really frightening way to look at any science.

      Delete
    3. Anonymous4:29 AM

      And yet I was taught that interest would be larger than loss of purchasing power rate. I was surprised interest has been so low for so long.

      Delete
  4. Anonymous5:22 PM

    don't focus on macro *theory*. It's macro empirics I'd worry about. theory ahead of measurement etc. The difference post crisis is you see greater prominence of macro papers using micro data (e.g. mian sufi or autor/dorn/hanson/acemoglu).

    ReplyDelete
    Replies
    1. Yes, great point.

      Delete
    2. But is this actually true? Davis/Haltiwanger/Hurst/whoever have been writing microdata macro papers since at least the 1990s.

      Delete
    3. With no data, it's a hypothesis and not a *theory.*

      Delete
    4. Anonymous11:34 AM

      @Ryan. Agree, people have been doing this work. I'm just guessing (hoping?) it's trendier now (obviously, an empirical question). More data, better computers and all that. Also, bleeding into public finance (chetty et al, etc) where you have administrative data (even Saez Piketty). For me, this is more interesting and worthwhile than calibrating one's favorite NK model.

      Delete
  5. On 1. in other parts of economics we have models about currency crises or public finance crises, and the perils of, say, excessive government debt are core ideas. There were of course models of banking crises, but they were peripheral and not paid much mind to. Predicting the crisis may have been too much to ask, but warnings about the danger could have been a core idea, and it wasn't.

    ReplyDelete
    Replies
    1. Yes, and I made comment at Mark Thoma's place (where I am semi-resident). I think this is the key point. I don't think expecting exact predictions is realistic (I actually think the world is stochastic). But plenty of people could see, and were warning (most prominently Dean Baker), that the situation was very dangerous. If the models were not capable of making this warning, then something very important was missing from them. (And how did they do in the aftermath - just how dynamic are these "dynamic" models.)

      Delete
  6. It wasn't just that mainstream macroeconomists and central bankers weren't focusing enough on finance. They combatively denied the importance of finance and rejected growing warnings from various quarters until it steamrollered them.

    ReplyDelete
    Replies
    1. And those who still cling to the fantasy that the business cycle is stochastic don't deserve the public funding they're still getting. Write off the sunk costs and move on already. Does it really cut it to just shrug and say, "well, nobody expects much from academic macro anyway, it's not that big of a portion of GDP anyway." So then, since $10b is only 0.06% of GDP, can I have that to stubbornly continue research on a disproved thesis?

      Delete
    2. Anonymous11:50 AM

      Simply repeating lies over and over again does not turn them into true statements.

      Delete
  7. Your point 1 is dubious.

    Take an example from finance. Before 2000, you had internet analysts justifying the price of Pets.com or whatever based on "eyeballs" or whatever crazy metric they could come up with.

    But one could have had a model/valuation metric that said that (a) Internet stocks are in a bubble, (b) it will end badly, and (c) there is no way to predict that when the "ending badly" will occur. Such an analysis is probably the best possible that could have been done, absent a time machine.

    But because the pessimist was unable to predict the exact timing of the peak in Internet stocks, your logic in (1) suggests that the pessimist's analysis was totally useless and therefore, the Internet analysts methodology was the best, since they at least had a quantitative prediction (stock prices go up by 50% a year!).

    What makes you think that any model can reliably predict recession timing? Once you drop the assumption that such a model exists, the existing heterodox modeling literature can be very useful. Wynne Godley's work is the most model-based of the post-Keynesians, if you want an example.

    ReplyDelete
    Replies
    1. "What makes you think that any model can reliably predict recession timing?"
      -Since the late 1960s, one based on the treasury yield curve. 100% of the time. Every time. But that tells us very little about severity. And it didn't work well between 1935 and 1967.

      Delete
    2. pithom - yes this is true, but only tells us there are lags in the economy (so the prediction is only identifying something that has in reality already happened). And predicting directions is not the same as predicting magnitude.

      Delete
  8. Anonymous7:32 PM

    Unfortunately economics like it is today is mostly an ideological propaganda and waste of time.
    The goal is ideological and the models used completely unrealistic and the logic completely flaw. It is like Galileo Galilei would use mathematical models to prove that earth is flat. They do not want to describe the world as it is but to build up a convenient vision. They can't describe as the the world and economy work, only an imaginative world.

    ReplyDelete
    Replies
    1. Anonymous4:08 AM

      Many a sailor and costal mountain hiker knew the world was not flat. Those people who see the world as it is.

      The shadow of the earth cast on the moon at a lunar eclipse is circular.

      There is a great story of the librarian, Eratosthenes, calculating the size of the earth using the angles of shadows cast in different places. (The easy way to understand it is to realize that the sun is so far away that the light and shadows fall on the earth in parallel. At all places on earth. And a little geometry.)

      The Indians have been so brilliant that I'm sure they knew too. The knowledge from India also must have been transmitted in contact with Greece and books ending up in the Alexandria library. The nile delta is very close to the Red Sea. Goods from the east could have gone easily over land to the delta and then to Alexandria and the Mediterranean sea. There was once an ancient canal connecting the Nile delta to the Red Sea.

      The European Dark ages was a local history. It ended at the renaissance and the age of "re"discovery.

      I mentioned a dodgy economic theory to some one and the quick reply was, "You don't have to believe that!"

      http://en.wikipedia.org/wiki/History_of_geodesy
      http://en.wikipedia.org/wiki/Eratosthenes
      http://www.windows2universe.org/citizen_science/myw/w2u_eratosthenes_calc_earth_size.html

      http://en.wikipedia.org/wiki/Indian_astronomy
      http://indiaheritage.org/science/astro.htm
      http://www.bharatvani.org/books/ait/ch22.htm
      http://www.starteachastronomy.com/indian.html
      http://www.indicethos.org/Astronomy/

      "The Dresden Codex [Mayan] contains astronomical tables of great accuracy. It is most famous for its Lunar Series and Venus table.[2] The lunar series has intervals correlating with eclipses. The Venus Table correlates with the apparent movements of the planet. The codex also contains almanacs, astronomical and astrological tables, and ritual schedules."
      ---http://en.wikipedia.org/wiki/Dresden_Codex#cite_note-ruggles1-2
      http://www.bibliotecapleyades.net/ciencia/dresden/dresdencodex03.htm
      http://www.tulane.edu/~danny/dresden.html
      http://ircamera.as.arizona.edu/NatSci102/NatSci102/text/extmayaastronomy.htm
      http://user.online.be/felixverbelen/cxd5158_1054web.pdf

      "The Science of Navigation: From Dead Reckoning to GPS", chapter 2, p. 34,36,, By Mark Denny
      https://books.google.com

      Delete
    2. "The European Dark ages was a local history. It ended at the renaissance and the age of "re"discovery."
      Oh God: historians don't use "Dark Ages" (it is propaganda, not history) and essentially no one in the Middle Ages thought the earth was flat! The roots of modern science were laid in the Middle Ages.

      Delete
    3. Yes, they do.
      https://books.google.com/books?id=VUy1RFS01yIC&printsec=frontcover&dq=%22dark+ages%22+rome&hl=en&sa=X&ei=M9QuVbiDJI21sQSS8IDoDQ&ved=0CB4Q6AEwAA#v=onepage&q=%22dark%20ages%22%20rome&f=false
      https://books.google.com/books?id=6GfOVMreQOwC&printsec=frontcover&dq=%22dark+ages%22+rome&hl=en&sa=X&ei=M9QuVbiDJI21sQSS8IDoDQ&ved=0CCgQ6AEwAg#v=onepage&q=%22dark%20ages%22%20rome&f=false
      https://books.google.com/books?id=s_2tNtNFZx4C&pg=PA76&dq=%22dark+ages%22+rome&hl=en&sa=X&ei=bNUuVerOG47egwTCloLYBw&ved=0CG0Q6AEwDg#v=onepage&q=%22dark%20ages%22%20rome&f=false
      The roots of modern science were laid before Christ.
      BTW, to all- Gene is a Catholic apologist.

      Delete
    4. Sorry for my last link: this is a better one.
      https://books.google.com/books?id=gmvbszxHFBwC&pg=PR15&dq=%22dark+ages%22+rome&hl=en&sa=X&ei=ReguVcL0O8mWyATXroHoBg&ved=0CDsQ6AEwBTh4#v=onepage&q=%22dark%20ages%22%20rome&f=false
      And, yes, the Dark Ages were real: Arab attacks led to the depopulation of Constantinople and Western Europe remained a cultural backwater until the High Middle Ages.

      Delete
    5. Anonymous3:23 AM

      "Unfortunately economics like it is today is mostly an ideological propaganda and waste of time.
      The goal is ideological and the models used completely unrealistic and the logic completely flaw. It is like Galileo Galilei would use mathematical models to prove that earth is flat. They do not want to describe the world as it is but to build up a convenient vision. They can't describe as the the world and economy work, only an imaginative world."

      +1 trillion

      Delete
    6. Anonymous4:26 AM

      Mr. Callahan,

      You are absolutely correct that few had any reason to think the world was flat and there was much observable evidence to the contrary.

      I should have specifically said this; many places outside the western world had excelled in sciences and mathematics for a large period of time before 1492 and Vasco de Gama's voyage. And, other foreign ideas probably have very well caused a large part of the Renaissance due to increased contact with the places and people the ideas and technologies came from. Indian Scholars of math and science history claim this.

      After all, the world is a lot bigger than Europe.

      I was, also implying a similar dark ages in Macroeconomics. And...

      -------------------------------------------------------------------------------------------
      Question (If you have studied calculus and calculus based any science): Do you think George Berkley might have made learning calculus harder?* I am referring to his critical views of the infinitesimal.

      The most difficult concepts taught in the calculus math class seem to address his criticisms of the "mathematical theoretical" foundations. Yet, none of it is in the physical sciences where the more clearer and more utilitarian ideas of Leibniz and Newton are used to great effect and clarity.

      I wonder if his ideas ruined the teaching calculus for students?

      * I saw you have been reading and writing about George Berkley.

      Delete
    7. Anonymous2:00 PM

      Continued:

      This is discussed on the George Berkeley wikipedia page in the section Philosophy of mathematics.
      http://en.wikipedia.org/wiki/George_Berkeley

      Some very small exerps
      " In 1734, he published The Analyst, subtitled A DISCOURSE Addressed to an Infidel Mathematician, a critique of the Calculus. Florian Cajori called this treatise "the most spectacular event of the century in the history of British mathematics."[25] However, a recent study suggests that Berkeley misunderstood Leibnizian calculus.[26] The mathematician in question is believed to have been either Edmond Halley, or Isaac Newton himself—though if to the latter, then the discourse was posthumously addressed, as Newton died in 1727. The Analyst represented a direct attack on the foundations and principles of calculus and, in particular, the notion of fluxion or infinitesimal change, which Newton and Leibniz used to develop the calculus. ..."

      "Berkeley regarded his criticism of calculus as part of his broader campaign against the religious implications of Newtonian mechanics – as a defence of traditional Christianity against deism, which tends to distance God from His worshipers. Specifically, he observed that both Newtonian and Leibnizian calculus employed infinitesimals sometimes as positive, nonzero quantities and other times as a number explicitly equal to zero. Berkeley's key point in "The Analyst" was that Newton's calculus (and the laws of motion based in calculus) lacked rigorous theoretical foundations."

      Comment: So, according to the above Berkeley was critical of the theoretical form calculus even though the method gave wonderful results in mechanics.

      He claimed that,

      "In every other Science Men prove their Conclusions by their Principles, and not their Principles by the Conclusions. But if in yours you should allow your selves this unnatural way of proceeding, the Consequence would be that you must take up with Induction, and bid adieu to Demonstration. And if you submit to this, your Authority will no longer lead the way in Points of Reason and Science.[27]"

      "Newton's science, therefore, could not on purely scientific grounds justify its conclusions, and the mechanical, deistic model of the universe could not be rationally justified."

      My comment is, math is an international language. Some times it is used as a very good description of nature. But, math does not make nature by fiat.

      Delete
    8. Anonymous2:10 PM

      The part of calculus that Berkeley was critical of is much easier to learn and apply to describe nature.

      Delete
  9. Anonymous8:49 PM

    Lot of hue and cry to state what many of us have long known through the lens of common sense - Keynesian economics doesn't work. It is simply a fig leaf for statist control.

    ReplyDelete
    Replies
    1. For the most part, agreed. Krugman and DeLong seem to be longing for a creeping government takeover of the economy, even if it doesn't boost growth.

      Delete
    2. ??????

      No. Look at what has actually happened and your thesis falls over.

      Delete
    3. P.S. I should note here (as Anonymous doesn't) that Keynes in fact wrote his most famous work in response to the Great Depression. And the tragedy now that we have a rerun is that to a large extent his policies recommendations were not followed so that the aftermath was worse than it should have been.

      Delete
    4. His policy recommendations were followed to the T. Deficit spending ballooned in every major economy.

      Delete
  10. "Macroeconomists don't do enough to kill their models off."

    I'm willing to assert that that, for every macroeconomist, there exists a plethora of models (or variations on models) that he or she would love to kill off. And plenty of people are swinging their swords with little effect. I'm not sure what more you expect us to be doing. 40 years ago (or thereabouts) some macroeconomists thought they had demonstrated unequivocally that the IS-LM model contained the seeds of its own destruction. But the thing lives on. This is a situation we're stuck with. Afraid you just have to let the flowers bloom and take your pick.

    ReplyDelete
    Replies
    1. I'm willing to assert that that, for every macroeconomist, there exists a plethora of models (or variations on models) that he or she would love to kill off. And plenty of people are swinging their swords with little effect. I'm not sure what more you expect us to be doing.

      Then tell me, why aren't the swords doing more cutting? Are macro models like those annoying enemies in Zelda that you can't hurt with your sword?

      You've played Zelda, right?

      Delete
    2. "Then tell me, why aren't the swords doing more cutting?"

      I gave you an example. Why don't you think that went away? By the way, I thought you were Mr. Behavioral Outside the Box. Seems you want to add some models. Think about that, and maybe you have your answer.

      Delete
    3. Other thoughts: A lot of the sword-swinging is done in private. There's no personal payoff from slaying dragons in public - that only makes enemies. There is, however, a payoff from thinking up a new idea - that's how we get tenure, get promoted, get good jobs, make money. An enemy won't give you a job. Some people are fearless, though - for example, Lucas, in his time. You can see that his still has enemies floating around, not to mention Noah Smiths who come along and misinterpret his motives 20 years later.

      Delete
    4. It appears to me that IS-LM has been applied quite usefully in the Great Recession. I am not aware of any other model that can say the same. I find it ironic that Stephen Williamson used it as an example of what should have disappeared.

      Delete
    5. Anonymous11:52 AM

      Your perception is incorrect. ISLM does not make predictions, because it is not an actual model -- it is a system of equations that, when estimated, will fit anything, There are no falsifiable implications, and therefore it is useless. We killed it off, or tried to, because it isn't a model.

      Delete
    6. Steve:

      A lot of the sword-swinging is done in private. There's no personal payoff from slaying dragons in public - that only makes enemies. There is, however, a payoff from thinking up a new idea - that's how we get tenure, get promoted, get good jobs, make money. An enemy won't give you a job.

      So what you're saying is that there's a perverse incentive system in place. In physics, all the theorists wait around hoping someone will test their theory and find out it works. Even the string theorists wish someone would find some way to verify their ideas.

      In econ, you're saying, theory gets full marks just for existing. So if anyone tries to bring it to data, they're challenging it, they're attacking it. They're an enemy.

      But is it really a good thing for models to get killed "in private"? Is that the right way for human knowledge to advance - through influential editors desk-rejecting models they don't like, or old guys talking bad about each others' students' papers behind closed doors?

      That doesn't sound like a good system to me.

      Delete
    7. Steve, let me follow that up with a question. If I started doing empirical tests of New Monetarist models, would you be happy or annoyed? Would you think "Awesome, now I get to see if my theory works"? Would you think "Awesome, finally some validation for my ideas"? Or would you think "That jerk, he's trying to kill my theory! Why doesn't he go make his own theories?"?

      Delete
    8. @Anonymous - My perception is that Paul Krugman made use of IS-LM to make a lot of very good predictions about how the economics of the Great Recession would play out.

      It's rather funny that you say IS-LM is not an "actual" model. Perhaps, you should argue with Stephen Williamson. It was his example of a model that should disappear, not mine.

      Delete
  11. Noah, you should really kill off Anonymous as an option.

    ReplyDelete
    Replies
    1. Anonymous10:44 PM

      Dunning Kruger ensures that plenty of people who might otherwise know better don't bother to be anonymous while embarrassing yourselves.

      Delete
    2. Ouuuuuccccccchhhhhhh

      The problem with Dunning-Kruger is that you don't actually get embarrassed. See, take me, I've never been embarrassed by anything... ;-)

      Delete
    3. Dunning Kruger?

      I never read that paper....

      Delete
    4. Anonymous6:35 PM

      I was going to make a joke about Megan McArdle here, but why bother.
      Instead, there's an enjoyable meta in your statement, Noah.

      Dear Edgar, only losers bring pedantry to a derp-fight.

      Delete
  12. Anonymous11:05 PM

    "Macroeconomists don't do enough to kill their models off."

    As a Macroeconomist this is exactly my complaint about Microeconomists!

    A counterargument does exist,
    https://ideas.repec.org/p/pen/papers/12-001.html
    not that I buy it.

    (I am more than happy for it to be equally applied to Macroeconomists)

    ReplyDelete
    Replies
    1. This argument is right as far as it goes, but a massive proliferation of crappy models will still make the field unusable.

      Delete
    2. Weird you say a proliferation of crappy models. As a macroeconomist I see too little diversity. Most of us (myself) included are in the DSGE camp tweaking the models to make them better. I wish there were more diversity. I believe in our ability to truncate the bad and go with the good models. In science variance is good. Truncation is the scientist's friend.

      Delete
  13. "Yeah, sure, if you put out hand-wavey reports saying "capitalism sux, there's gonna be a crash!" every year or two, you're eventually going to be able to say "see, I told you so". But that's no replacement for real modeling."

    Wow! what dishonesty Noah.

    First it's not chart blogging and you are lying.

    Second the paper you refer to doesn't say capitalism sucks. It talks of unsustainable processes which need to be corrected by policy.

    But else can one expect from toadies.

    ReplyDelete
  14. Anonymous2:16 AM

    How about.....

    Macro economic theories are so powerful that they caused the crisis, but they didn't predict it.

    ReplyDelete
  15. Anonymous3:14 AM

    "When I ask angry "heterodox" people "what better alternative models are there?", they usually either mention some models but fail to provide links and then quickly change the subject, or they link me to reports that are basically just chartblogging."

    I am glad you put heterodox in inverted commas because I do not think anybody here has a clue who are you talking about. Are you talking about critics of the mainstream profession, or the likes of Angus Maddison and Leijonhuvfud? Are you talking about people in the tradition of Keynes, Robinson and Kaldor? And what do you mean by "have a model"? Many people are historians who for very good reasons do not "have a 'model' ". Are you talking about a coherent explanation of an historical case? Are you talking about a universally applicable theory that applies to all cases?

    I have a feeling who are targeting a group and misrepresenting them to get out of another battle. And the anger if there is anyway is about poverty and injustice. But to represent them all, or probably any of them, as hot-headed reactionaries is not helpful for solving anything. There are big problems in economies at the moment, and arguably unless a fundamentally flawed discipline is put under scrutiny, it will remain useless in trying to solve them.

    ReplyDelete
    Replies
    1. Anonymous5:44 AM

      "and arguably unless a fundamentally flawed discipline is put under scrutiny, it will remain useless in trying to solve them."

      I think any solutions are most likely to come from the outside. From people versed in business, accounting, and fields that require actual experimental lab classes and system analysis and design such as most engineering.

      Why accounting? It is quantitative measurement and analysis that can take into account all types of transactions. Check out Dirk Bezimer's paper on Accounting models. The best part is, many economists on line act like they don't need to know any of it or just basterdize accounting badly. So, improvement is more likely to come from the outside and some economists that are willing to reach out to other fields. It is going to be easier for the outsiders.

      Delete
    2. Anonymous11:55 AM

      Gods you're dumb.

      Delete
  16. I agree on killing bad models. I suggest we start with anything based on the Permanent Income Hypothesis, NAIRU, and expansionary fiscal austerity.
    The rest, is just an assembly of contradictory and incoherent thoughts, both with the article itself, and with other things you've written. I think you can do better than this...

    ReplyDelete
    Replies
    1. PIH and NAIRU are correct. Expansionary fiscal austerity can happen in certain limited cases, especially if it helps the supply-side.

      Delete
    2. Sorry to break the news to you, but PIH has been thoroughly refuted, the NAIRU as well (even by strong adherents to that religious creed such as the ECB and the IMF), the same for expansionary fiscal austerity. If there's any standard for considered a theory as refuted, those 3 fit the requirements.

      Delete
    3. [citation needed].

      Delete
    4. On expansionary fiscal austerity: https://www.imf.org/external/pubs/ft/wp/2011/wp11158.pdf

      On NAIRU: IMF World Economic Outlook chapt 3, and ECB Working Paper 1763.
      On PIH, well, where to start? I give you just one. There are A LOT by Robert Frank. https://www.aeaweb.org/annual_mtg_papers/2007/0107_1300_0202.pdf

      Delete
    5. And in this post you'll find many more links to educate yourself.
      https://unlearningeconomics.wordpress.com/tag/the-permanent-income-hypothesis/

      Delete
    6. "The model thus predicts that savings rates should be independent of
      household income and should remain stable over time"
      -Never heard Milton Friedman say anything like this. This may be a result of my ignorance, but I'd bet against it if I had a dollar to waste.

      Delete
    7. Also, that unlearning economics link had a clear bit of nonsense:
      "If inflation is purely monetary, then the level of unemployment should not affect it at all!"
      -This clearly shows that that bit of nonsense's writer had no idea what Friedman meant. Friedman obviously meant that unemployment doesn't affect nominal labor demand, rather, nominal labor demand affects unemployment, and there is a level of unemployment (which may be, as in the 1970s, rather high) which no amount of nominal labor demand will be able to lower.
      This is why the best-looking Phillips curve uses average hourly earnings instead of inflation. Inflation works, too, BTW.

      No wonder the site is called "unlearning economics". The author seems to have forgotten a good chunk of Econ 101.

      Delete
    8. More evidence the author doesn't understand Friedman:
      "This always struck me as strange – he was, in effect, arguing that the Great Depression was the fault of ‘the government’ because they failed to intervene sufficiently. This implies that the real source of the Great Depression came from somewhere other than the Federal Reserve, and therefore its sin was more one of omission than commission. Even if we accept the idea that the Great Depression was worsened by the action (or inaction) of central banks, Friedman is being disingenuous when he says that the Great Depression was “produced” by the government."
      -As webpages from the now sadly defunct decouple.org show, Friedman's view was much more sophisticated than the author posits. Friedman argued that the establishment of the Fed fundamentally changed how banks respond to crises, which necessitated that, while the Fed was previously unnecessary, its existence necessitated that it do QE in case of Great Contraction-style Aggregate Demand shocks.

      Delete
    9. Third strike:
      "Bottom line? Friedman had to assume his conclusion – that the money supply was in control of the Federal Reserve – in order to reach it."
      -No, all he had to do was look at the Fed's powers.
      Fourth strike:
      "Yet, based on his own numbers, his conclusion was still false, as the money supply increased over the ‘crash’ period from 1929-1931."
      -Base money is one thing, M1 is another thing entirely.
      No thanks for linking me to an ignoramus who unlearned a whole bunch of econ, PGB.

      Delete
    10. Also, on expansionary austerity, the paper you referenced doesn't adjust for the effects of monetary offset.

      Delete
    11. See for yourself the amount of REFUTED assumptions and corollaries of the Permanent Income Hypothesis: http://www.nytimes.com/2005/06/09/business/09scene.html?_r=0


      The flat segment of the Phillips curve, which is now an accepted empirical regularity (besides the IMF and ECB papers that I mentioned, a NY Fed paper by Peach, Rich & Cororaton 2011, Stock & Watson 2010, Barnes & Olivei of the Boston Fed 2003) denies the existence of A NAIRU. But we already know that you are impervious to the real world, I gave up on expecting you to understand it.

      Delete
    12. Since you responded to none of my points, I feel no obligation respond to any of yours. So you think you can lower unemployment to zero by simply increasing private sector nominal labor demand?

      Delete
    13. It's ridiculous to lower it to "zero". For all practical purposes, 2% is full-employment, as the postwar years show. Besides that, there can be some supply constraints like lack of productive capacity (though in normal economies, that limit is hardly ever reached). For developing countries, the most relevant constraint is the foreign constraint, lack of foreign currencies, which sets a limit to the expansion of real aggregate demand. But investment is demand-led, no question about it. Even the IMF (traditional champion of supply-side policies) recognizes it. Hard to understand why you don't.

      Delete
    14. "For all practical purposes, 2% is full-employment, as the postwar years show."
      -I thought of writing I thought the NAIRU was about 3-4% today, but I didn't do so. As the 1970s show, productive capacity may be substantially idle in one sector while at full employment in others, with underproduction in only a few sectors holding the rest of the economy down.
      "For developing countries, the most relevant constraint is the foreign constraint, lack of foreign currencies, which sets a limit to the expansion of real aggregate demand."
      -Definitely. Brazil in the 1980s is a great example of this.
      What, exactly, do you mean by "demand-led" in your comment?

      Delete
  17. Anonymous10:02 AM

    Nikki Kaldor for example tried to describe the world as it is and to understand it.
    Economics is a soft social science which has to deal a lot with the power too, so the right beginning should be put down the right logic.
    The mainstream wants to hide it, and modeling turns to be an ideological exercise. They build up a Fugazi world.

    ReplyDelete
  18. Apropos "macroeconomists. . .take their models way too seriously":

    "I have yet to meet a successful scientist who lacks the ability to exaggerate the importance of what he or she is doing." - Daniel Kahneman

    ReplyDelete
  19. Hmmm. I do not want to revisit the "who predicted the crash and how?" debate, but will add to those who say that the linked-to-model is certainly a bit more than mere "chartblogging." I would also say that not all the hets who claimed to have made more or less accurate forecasts based on more or less viable models are "angry." Some at least are just frustrated.

    I would also agree that we do not have a clear way to eliminate models unequivocally as "wrong."

    Barkley Rosser

    ReplyDelete
  20. I will add regarding the Fed that apparently the first person there in a position of real power and not just some low level researcher hiding in a basement closet somewhere to have become worried that the housing bubble in the US was a serious problem and threat to the US economy was none other than Janet Yellen, with the WSJ identifying her as more broadly having the best forecasting record of major players at the Fed, whatever the foundation or basis of those forecasts. Which I think is a good reason to be grateful that she is now Chair of Fed Board.

    JBR

    ReplyDelete
    Replies
    1. "was a serious problem and threat to the US economy was none other than Janet Yellen"
      -True-in the second half of 2007.
      In the first half, she was still saying
      "While a tightening of credit to the subprime sector and foreclosures on existing properties have the potential to deepen the housing downturn, I do not consider it very likely that such developments will have a big effect on overall U.S. economic performance. I say this, in part, because these mortgages represent only a small part of the overall outstanding mortgage stock."
      -I wonder if she still agrees with that statement.

      Delete
    2. She initially expressed concern about the housing bubble in 2005.

      Delete
  21. "Now, technically, you could choose between models based on the plausibility of the assumptions..."

    Well, guess what. Human beings can be deceptive. Actually, some of what Matt Rognlie does is not his specialty. That's what we're all doing - constantly. That's why we like the models to be explicit - microfounded as it were. That's not just about Lucas critique. You have to be able to take it apart and see how it works. But of course that's work for someone with some technical expertise. It's not going to be obvious to the average blogger.

    ReplyDelete
    Replies
    1. Actually, some of what Matt Rognlie does is not his specialty.

      What does that mean?

      Delete
    2. He's not the only one doing it?

      Delete
  22. I am less concerned with an inability to predict the crisis than the inability to coherently promote economic policy in the aftermath.

    Economists still can't agree if fiscal policy works. That's pathetic.

    ReplyDelete
    Replies
    1. The vast majority of everyday PhD economists agree fiscal policy works. As Sumner says, fiscal policy only works if a sovereign monetary policy is unavailable.

      Delete
    2. This to me is the key point. I don't fault anybody for being unable to predict the crash. More to the point is what did and didn't happen afterwards.

      Delete
    3. @Pithom -- You may know better, but I think there's a lot of voices, Cochrane being one example, that publicly say fiscal policy doesn't work. There's others, Krugman being the main advocate, saying fiscal policy works at the ZLB. Sumner appears to me to be a third voice saying NGDP targeting makes the ZLB a non-issue, so fiscal policy is unnecessary.

      Delete
  23. And, is anyone really surprised that classically based models were blind to depressions?

    ReplyDelete
  24. Economists who are not macroeconomists generally have a pretty jaundiced view of macro and always have. There are some big ideas you can take to the bank: productivity growth matters a lot, so does the business cycle, monetary growth, and aggregate demand. But otherwise Yorum Bauman pretty much got it right when he parodied the “ten principles of economics” in Greg Mankiw’s best-selling textbook in the Annals of Improbable Research (2003): blah, blah, blah. Frankly, it can hardly be otherwise, we have an overdetermined system with a paucity of data points, crude mechanisms, and no ability run controlled experiments.

    ReplyDelete
  25. DSGE sucks because the model becomes slave to the tractability of the math.

    In my view it is fair to lump all of economists together. If you are not exposing the charlatans among you then it is fair for the rest of us to tar you all with the same brush. I would consider the "charlatans" to include the economists who provided intellectual cover for the deregulation of the financial markets and anyone who has expressed or endorsed the view that the housing bubble and financial crisis were caused by the Community Reinvestment Act (this catches everyone associated with the Romney presidential campaign.)

    The one economist who seems to have expressed real concern about the financial markets is Raghuram Rajan http://www.nber.org/papers/w11728

    ReplyDelete
    Replies
    1. What's wrong with deregulation? Anybody who thinks the CRA caused "the housing bubble" is a shockingly lazy thinker. Housing bubbles appeared in 2001-7 in Spain, the Baltics, the Netherlands, Denmark, Ireland, and South Africa, and there were sustainable house price increases in Belgium, Australia, Canada, France, and Sweden at precisely the same time.
      http://www.economist.com/blogs/dailychart/2011/11/global-house-prices

      Delete
    2. Rajan stuck his neck out early at a Jackson Hole meeting of Fed economists, but there were other economists who got it pretty well before he did, and parts of his argument were problematic.

      One can argue that he did not use a sophisticated model to do it, but it is widely accepted (and I think that Noah agrees) that probably Dean Baker was the first economist to call what was coming fairly fully and not in some ridiculous way like so many who were forecasting doom constantly any minute for years.

      In short, while Rajan did pretty well, he was hardly the first.

      Delete
  26. "Belief in the Great Moderation, and in the Fed's ability to stabilize the economy, was too strong."

    If I may (perhaps inaccurately) attempt to channel Sumner, do we know that? Or did the Fed simply fail to act as it could have to stabilize the economy in 2008?

    ReplyDelete
  27. How does one model a false premise?

    "Housing prices will not fall" was a premise for the excessive risk taking (on the part of finance and consumers) that lead directly to the crisis.

    Many models must predict the crisis if you build that as a premise and suddenly change it.

    So any "predicting" model would essentially have to weigh the probabilities of any of its own premeses being false and play the scenario of negating them - one at a time or several at a time. Then the model explodes, and you have a crisis prediction.
    The value of the prediction is dubious.

    It seems like the only meaningful predictive model of the crises would be one that explicitly foretold the collapse of home prices.

    You might want to talk to Mr. Roubini on that one.

    ReplyDelete
    Replies
    1. Max, why didn't housing prices fall in our neighbor to the North? And why did they fall here? I don't know the answer, so these are not rhetorical questions.

      Delete
    2. I don't see it.
      http://www.statcan.gc.ca/pub/91-003-x/2007001/figures/figure17-en.gif
      Why weren't Canadian housing prices higher before 2001?

      Delete
    3. Max, why didn't housing prices fall in our neighbor to the North?

      (1) High oil prices.
      (2) A better regulated financial sector and a government willing to step in and provide liquidity support to the banks for mortgage lending without a lot of fuss and delay.

      Delete
    4. @Absalon
      -High oil prices might explain things in early 2008. But not in late 2008. And what hath oil to do with housing?
      Canada had a better-regulated financial sector, but the same housing price increase the U.S. had, as by the Economist blog chart I linked to above. Why was that?

      Delete
  28. Dr. Smith,

    Lots of people foresaw the Housing Bubble bursting and imagined the broader economic consequences.


    April 2006, CBS News


    "Countless articles in the financial and popular press have now been devoted to the question of whether we are in a housing 'bubble.' It is a favorite topic of many liberal economists, columnists, and bloggers, who argue that President Bush's tax cuts and other policies have created a hollow and unsustainable economy. They are laying the groundwork to hang a housing bust around the necks of President Bush and congressional Republicans."[1]


    May 2002, Los Angeles Times


    "After all the rhapsody about the digital age, it was skyrocketing home prices and sales that finally shook the U.S. economy out of its doldrums this year. While stock markets languished, mortgage refinancing pumped more than $1 trillion into consumers' pockets. Nationally, average home prices rose above the $200,000 mark for the first time ever. But rather than signal prosperity, the housing boom may be largely driven by the financial maneuvers of wealthier Americans. Housing, in fact, has become part of the bubble economy, with potential consequences more serious than the stock market fiasco."[2]


    September 3, 2001


    "The downside to easy credit: overleverage. Most households have either no mortgages or very manageable ones. But there are enough people stretching their budgets to have caused a key debt-burden ratio to hit an alltime high. The ratio of mortgage debt service to total disposable income climbed to 6.46% in the fourth quarter of 2000, surpassing a 6.35% record set during the first quarter of 1991 in the depth of the last recession. Collective owners’ equity in the U.S., as a percentage of the real estate’s value, sank to 55% in the first quarter of this year, the lowest level ever and down from 70% in 1982. 'Leverage against an asset that can deflate in value is a recipe for disaster,' says economist Charles W. Peabody of Mitchell Securities in Manhattan."[3]


    Anyone who *wanted* to see the problems, saw them. Those who did not see the crash coming, did not want to see it coming. The problem was that too many people had a vested interest in not seeing the problem. It was not a theoretical or analytical problem but a political problem.


    [1] http://cbsn.ws/12A5bLP
    [2] http://lat.ms/Y1suwe
    [3] http://onforb.es/Y1ssVb

    ReplyDelete
  29. Anonymous1:45 PM

    When DA says

    "The point I want to make here is not that the DGE approach is the only way to go. I am not saying this at all. In fact, I personally believe in the coexistence of many different methodologies. The science of economics is not settled, after all. The point I am trying to make is that the DGE approach is not insensible ."

    He is being utterly disingenuous. In fact, he is lying.

    ReplyDelete
  30. As a Computer Scientist who happened to get an Econ degree, my (probably incorrect) impression is that most of the economics profession is in denial when it comes to General Equilibrium Theory which lead to an unhealthy faith in the Great Moderation.

    Back in grad school, when I tried to actually implement an distributed agent market for power allocation in a destroyer, I probably consulted 2 dozen Econ professors. 95% were completely useless as in all of their work they just assumed General Equilibrium "happened" and instantly. And when the system wasn't converging to global optimum... silence.

    It should be obvious, if these models of agents (at least the models that don't assume the agents somehow already know the global optimum) actually could find global optimums consistently, non-linear programming would be done.

    You guys should ditch "markets" and utility optimizing agents in favor of simulated annealing. It's a better algo.

    ReplyDelete
    Replies
    1. Anonymous2:44 PM

      You clearly didn't learn anything from your Econ "degree".

      Delete
  31. thanks info guys, "Housing prices will not fall" was a premise for the excessive risk taking (on the part of finance and consumers) that lead directly to the crisis.

    Many models must predict the crisis if you build that as a premise and suddenly change it.

    ReplyDelete
  32. Good work and nice Post, I highly appreciate those people who share some good information.Thanks for sharing.
    Shahid Nazir Ahmad

    ReplyDelete
  33. "Macro models failed to predict the crisis, therefore DSGE sucks." - what about this: all of economics is backwards looking, nothing but data fitting, and the entire profession is snake oil? There's no 'predicting' a crisis anymore than you can 'predict' when a sand pile will collapse. It's all nonlinear. At best you can outline broad parameters, as they do with climate (100 year floods, etc).

    ReplyDelete