Some random thoughts on modern macro theory

After returning from a family vacation in Mexico (aka a week long upset stomach I’m still trying to recover from) I have been trying to catch up on the blogosphere.  Tyler Cowen directed me to a series of interesting posts by JW Mason, John Cochrane, (and indirectly) Noah Smith.  All these individuals know much more modern macro than I do, but I’ll provide my two cents worth anyway.  My basic approach is as follows:

1.  Like Milton Friedman, I prefer to use simplest model for the problem at hand.

2.  I prefer to use different models for those business cycles that are created by “nominal shocks” (defined below) and everything else, i.e. “real shocks.”  Real shocks include factors that change the natural rate of unemployment (or LFPR), savings preferences, technology shocks, terms of trade, taxes, government output, etc.  All these things would matter even if wages and prices were completely flexible.  Indeed real shocks would be all that matters.  Indeed they are (almost) all that matters in the long run.

3.  The problem of nominal shocks combined with sticky wages/prices is so radically different from the rest of economics that attempts to capture all of macro in a single model leads to DSGE models that simply cannot do what we ask of them—provide reliable policy advice.  Here’s Noah Smith:

Imagine a huge supermarket isle a kilometer long, packed with a million different kinds of peanut butter. And imagine that all the peanut butter brands look very similar, with the differences relegated to the ingredients lists on the back, which are all things like “potassium benzoate”. Now imagine that 85% of the peanut butter brands are actually poisonous, and that only a sophisticated understanding of the chemistry of things like potassium benzoate will allow you to tell which are good and which are poisonous.

This scenario, I think, gives a good general description of the problem facing any policymaker who wants to take DSGE models at face value and use them to inform government policy.

Read the whole thing.

4.  My preferred nominal shock model would focus on explaining movements in nominal aggregates, and also explaining the business cycle.  More specifically, explaining that part of the business cycle that is due to employment fluctuations attributable to wage and price stickiness.  My hunch is that wage stickiness is the key, so I’ll focus on that issue.

5.  Because of wage/price stickiness, we need some sort of indicator of “nominal shocks.”  We do not have such an indicator, although I often talk as if NGDP is the optimal nominal indicator.  Let’s call the optimal nominal indicator “ONI,” and define it as the variable that, if stabilized by the Fed, would leave employment levels closest to their flexible wage equilibrium, i.e. at a level that would obtain if all wages (and prices) were completely flexible.  Thus “nominal shocks” can only be defined in terms of wage stickiness, and the ONI will depend on the precise type of wage stickiness.  Nominal shocks are not self-evident like “oil price shocks”—you need a model.  NGDP is almost certainly not the ONI, as there are presumably monetary policies that would keep employment closer to its flexible wage level than a NGDPLT policy regime.

6.  The term ‘demand’ is a rather unfortunate carry-over from the Keynesian revolution.  Here’s Mason:

People often talk about aggregate demand as if it were a quantity. But this is not exactly right. There’s no number or set of numbers in the national accounts labeled “aggregate demand”. Rather, aggregate demand is a way of interpreting the numbers in the national accounts. (Admittedly, it’s the way of interpreting them that guided their creation in the first place). It’s a statement about a relationship between economic quantities. Specifically, it’s a statement that we should think about current income and current expenditure as mutually determining each other.

I use the term ‘demand’ in order to talk to Keynesians.  However when I use ‘demand’ I have something in mind that is a specific quantity (NGDP, or ONI), and thus I am not really speaking the Keynesian language—I’m talking a close dialect (like Cantonese to Mandarin.)  I’d like macroeconomists to stop talking about “demand” and start talking about their preferred ONI.

7.  It seems to me that most macro models don’t have a ONI, but rather contain factors that might or might not affect the true ONI (i.e. interest rates, fiscal policy, etc.)  And I think that’s part of the problem.  Here’s Cochrane:

All current macroeconomic theories start with the same basic story: when interest rates are higher, people consume less today, save, and then consume more in the future.

Why isn’t this an example of the fallacy of “reasoning from a price change?”  Cochrane clearly understands this issue because in subsequent paragraphs he talks about the NK view that an outbreak of “thrift” triggered both much lower interest rates and much lower consumption.  Aren’t we asking too much of interest rates?  They are a key variable for intertemporal allocation decisions, but the NK models also use them to represent monetary policy.  That’s just one of the reasons I prefer to use an ONI as an indicator of monetary policy.

8.  Later Cochrane discusses the NK view that inflation, even supply-side inflation, can boost real consumption:

Fiscal stimulus, and many of the other seemingly magical properties of new-Keynesian models (see  last post) follow from the idea that inflation is good. Fiscal stimulus raises inflation. Broken windows, hurricanes, pointless public works projects, temporarily lowering the economy’s productive capacity, all raise inflation (how is in other equations of the model), which lowers interest rates.

I’m not sold on this story, as you probably guessed, for a variety of reasons.

I’m also skeptical, but for reasons that Cochrane would probably reject even more vehemently than the NK models he criticizes. The view that disasters boost output violates the undergrad AS/AD approach to macro, which predicts that a hurricane will shift AS to the left, reducing output.  I believe the textbook AS/AD modelers stumbled on something close to the truth, for essentially accidental reasons.  The AD curve is usually drawn with a curvature that looks vaguely like a hyperbola, but not exactly.  That means “demand” is assumed to be loosely related to P*Y, or NGDP.  In other words, “demand” is a sort of loose proxy for the ONI.  There is no actual “ONI” in Keynesian models, but the simple undergrad AD curve is close enough that it produces pretty decent empirical results, indeed better empirical results than NK models featuring weird upward sloping AD curves.

10.  If you insist on a formal model, call the natural rate of employment (in terms of hours worked) “En.”  Then E – En is negatively related to W/ONI, and hourly wages are very “sticky” in the short run, especially when the aggregate rate of hourly wage rate growth approaches zero.  Thus if the central bank can control ONI, it can also influence E in the short run.  The musical chairs model.  The rest of the economy can be modeled using new classical principles.

PS.  It’s probably best to visualize my simple model by using “aggregate wage and salary income” as the ONI, rather than NGDP.

PPS.  I recommend two new Nick Rowe posts on NGDP targeting and the Phillips Curve.

PPS.  I’ll read all the old comments, and answer as many (or few) as I have the energy for.

PPPS.  Did they really close down Boston when I was gone?  If so, why?  Or to me more specific, can someone write down a rule that determines when the authorities should close down large cities?  I’m not trying to be cute, I actually don’t understand the rationale, although I could imagine several different possible explanations.  But I’d be more interested in what you think.  I’m inclined to see this as a harbinger of things to come.  Welcome to the 21st century.


Tags:

 
 
 

53 Responses to “Some random thoughts on modern macro theory”

  1. Gravatar of Greg Ransom Greg Ransom
    22. April 2013 at 09:34

    Here’s the key.

    Macroeconomists are like Ptolomaic astronomist/astrologists who found it simplest and most in harmony with their background magical thinking and religious beliefs to view the cosmos as simple sets of nested perfect spheres, with the earth, very simply, the center of the universe.

    Like with contemporary macroeconomists, getting the reality of the structure of the world right didn’t matter to these simple-minded, magically thinking men.

    Case in point — contemporary macroeconomists don’t want to get the time structure of the coordination of production goods across time right — they prefer a “simply” and magical conception which patently conflicts with the causal structure of the real world.

  2. Gravatar of Geoff Geoff
    22. April 2013 at 10:03

    Real shocks cause nominal shocks and nominal shocks cause real shocks. They don’t occur in isolation, divorced from the other.

    There are real reasons why people suddenly decide to spend less, and there are nominal reasons why people suddenly decide to produce less.

    Obliterating market determined fluctuations in NGDP at the national level obliterates the stabilizing effects such market determined fluctuations has on inter-national, i.e. world, market activity.

  3. Gravatar of Tyler Joyner Tyler Joyner
    22. April 2013 at 10:14

    “Did they really close down Boston when I was gone? If so, why? Or to me more specific, can someone write down a rule that determines when the authorities should close down large cities?”

    My guess would be that they were not yet certain whether the bombing was an isolated incident, or if more attacks in and around Boston were impending. There was the confusion with the JFK library, initial reports of undetonated bombs around the marathon, etc. Of course the general militarization of our police force and other consequences of the never-ending war on terrorism also apply.

  4. Gravatar of ssumner ssumner
    22. April 2013 at 10:20

    Tyler, I’m not so sure, I was under the impression that the shutdown didn’t happen until about 4 days after the bombing–hence I doubt it was simply fear of additional bombings, although obviously that might be one explanation. Indeed by Friday one of the bombers was dead and the other was wounded and on the run, so it would seem the likelihood of another Marathon-type bombing had declined.

  5. Gravatar of Tyler Joyner Tyler Joyner
    22. April 2013 at 10:24

    Hmm, I must have my timeline of events confused.

    I remember one time years ago driving through LA when someone in the city shot a police officer. The highways were a parking lot for 5-6 hours while they shut everything down so they could catch the guy.

    Maybe in Boston they had imperfect intel on the location of the wounded guy, and thought he might shoot people randomly if he encountered them? Who knows.

  6. Gravatar of Doug M Doug M
    22. April 2013 at 10:45

    I don’t see the value in separating real shocks from nominal shocks. Was the 2008 financial collapse a real event or a nominal event? I know the professor’s opinion, but I think that many would disagree.

    Regarding “aggregage demand”, I am more than ready to scrap this term from the vocabulary of economics.

  7. Gravatar of marcus nunes marcus nunes
    22. April 2013 at 11:03

    Scott
    While you were ‘vacationing’ Miles Kimball argued that (positive) productivity shocks are contractionary. I countered:
    http://thefaintofheart.wordpress.com/2013/04/21/are-technology-shocks-contractionary-miles-kimballs-conjecture/

  8. Gravatar of marcus nunes marcus nunes
    22. April 2013 at 11:05

    If you needed proof that the world spins around…
    http://thefaintofheart.wordpress.com/2013/04/22/rogoff-should-have-remembered-his-2002-letter-to-joe-stiglitz/

  9. Gravatar of Steve Steve
    22. April 2013 at 11:08

    Pictures of Boston shutdown:

    http://www.theatlanticwire.com/national/2013/04/boston-lockdown-residents-are-asked-shelter-place-while-cops-sweep-watertown/64383/

    The presumed rationale is that the at-large terrorists had more bombs (true) and would detonate them in a crowd (e.g. red line subway) if they could. Plus the lack of traffic makes it hard to escape by changing clothes, carjacking, etc. You could debate the extent of the shutdown however.

  10. Gravatar of Bill Ellis Bill Ellis
    22. April 2013 at 11:14

    “Did they really close down Boston when I was gone? ”

    When you go on vacation… you really go on vacation. Good for you.

  11. Gravatar of Integral Integral
    22. April 2013 at 11:20

    Scott,

    Yeah, they shut down Boston for a couple of hours on Friday (roughly, 7am to 6pm). “Shelter-in-place” was under effect for Boson, Newton, Watertown, and a few other areas, during the manhunt. The T was closed down and cab service was suspended.

    On the ONI: interestingly enough, in a wide, wide, wide class of New Keynesian models, P is the ONI. Stabilizing P allows the central bank to mimic the flexible-price equilibrium and restore the model’s underlying, Pareto-optimal RBC dynamics. This “works” for labor productivity shocks, for government spending shocks, and for monetary shocks.

  12. Gravatar of Ashok Rao Ashok Rao
    22. April 2013 at 11:41

    Scott,

    I live in India, but Boston was home for some years before we moved here for high school. Seeing pictures of tanks rolling down the streets – for, what, a runaway 19 year old – scared me. Only conclusion I reach is this is the first time, since 9/11, America has a chance of trying its new security systems.

    It’s a show of strength, to the country and world. I think.

    When I first read your post, I thought it would be cute little remarks on modern MONETARY theory! 😉

    What is your opinion on computational or agent-based models? I don’t say they are a panacea, but we don’t know that they don’t work which, to me, is a fantastic reason to try. Europe is ahead of us on this, and I don’t know why the Fed can’t spend some more $ buying a supercomputer.

  13. Gravatar of marcus nunes marcus nunes
    22. April 2013 at 11:50

    @Ashok
    The people at the Santa Fe Insitute are ‘deep’ into that sort of modelling.

  14. Gravatar of Don Geddis Don Geddis
    22. April 2013 at 11:51

    @Doug: “I don’t see the value in separating real shocks from nominal shocks.

    Surely you can accept that one huge difference between the two, is that government has limited powers to ameliorate real shocks. But the central bank has essentially infinite power to set pretty much any nominal level that it wants. Don’t you find that makes it valuable to analyze the two kinds of shocks separately?

  15. Gravatar of Ashok Rao Ashok Rao
    22. April 2013 at 11:56

    Marcus,

    I’m an intern there – which is why I’m particularly interested. I’ll be in college next year, and my degree program is very similar to some of the complexity science stuff.

    I’ve written quite a bit about this on my blog.

  16. Gravatar of marcus nunes marcus nunes
    22. April 2013 at 12:04

    @Ashok
    Talk about ‘teaching the gospel to St Peter’! Hope you are enjoying your internship. I´ve just finished readin a nice little book by Mark Buchanan very critical of the E in DSGE:
    http://www.amazon.com/Forecast-Physics-Meteorology-Sciences-Economics/dp/1608198510/ref=sr_1_1?s=books&ie=UTF8&qid=1366660916&sr=1-1&keywords=forecast

  17. Gravatar of Steve Steve
    22. April 2013 at 12:13

    Tim Duy has a rambling commentary on the “monetary trilemma”. Apparently he doesn’t read your blog.

    http://economistsview.typepad.com/timduy/2013/04/monetary-policy-and-financial-stability.html

  18. Gravatar of Ashok Rao Ashok Rao
    22. April 2013 at 12:16

    Marcus, HARDLY the right analogy, way overstates my knowledge! My work is also in a different area (scaling of cities) but definitely fascinating. (

    Yes, Buchanan has been very critical of “elegant models” and a great op-ed out recently, http://www.bloomberg.com/news/2013-04-07/beware-of-economists-peddling-elegant-models.html

    If this sort of thing interests you, I’ll direct you to this new paper by W Brian Arthur, a stalwart of the field, though the book you’re reading might have talked all about it:

    http://www.santafe.edu/media/workingpapers/13-04-012.pdf

    The broader theme is definitely questioning the “E” in DSGE.

  19. Gravatar of marcus nunes marcus nunes
    22. April 2013 at 12:23

    @Ashok
    Thanks for the tips.

  20. Gravatar of J J
    22. April 2013 at 12:31

    On Boston closing down:

    Once the word ‘terrorist’ comes out, there is no way that the government can allow him/her (has it ever been a her recently?) to escape. I’m not saying this is logical; this may just be a political move because it looks bad to let a terrorist get away. Most of the news reports at the time suggested that closing everything down was to allow the police easy access to wherever they wanted to go and to allow gun fights or whatever kind of force was necessary without fear of collateral damage.

  21. Gravatar of marcus nunes marcus nunes
    22. April 2013 at 12:47

    @Ashok
    You might be interested in this old “Krugman attack on Brian Arthur”:
    http://www.pkarchive.org/

  22. Gravatar of marcus nunes marcus nunes
    22. April 2013 at 12:48

    @Ashok
    You may be interested in this old “Krugman attack on Brian Arthur”:
    http://www.pkarchive.org/

  23. Gravatar of marcus nunes marcus nunes
    22. April 2013 at 12:55

    @Ashok
    The link is not direct. On the left bar click on ‘Cranks’ and then scroll to ‘Legend of Arthur’ in 1.98

  24. Gravatar of Ashok Rao Ashok Rao
    22. April 2013 at 13:01

    Got it. Anyway, I think that was an unfair attack by Krugman. Arrow destroys the argument:

    “””Paul Krugman’s attack on Brian Arthur (“The Legend of Arthur”) requires a correction of its misrepresentations of fact. Arthur is a reputable and significant scholar whose work is indeed having influence in the field of industrial organization and in particular public policy toward antitrust policy in hightech industries. Krugman admits that he wrote the article because he was “just pissed off,” not a very good state for a judicious statement of facts, as his column shows.

    His theme is stated in his first paragraph: “Cassidy’s article [in The New Yorker of Jan. 12] tells the story of how Stanford Professor Brian Arthur came up with the idea of increasing returns.” Cassidy, however, said nothing of the sort. The concept of increasing returns is indeed very old, and Cassidy at no point attributed that idea to Arthur. Indeed, the phrase “increasing returns” appears just once in Cassidy’s article and then merely to say that Arthur had used the term while others refer to network externalities. Further, Arthur has never made any such preposterous claim at any other time. On the contrary, his papers have fully cited the history of the field and made references to the previous papers, including those of Paul Krugman. (See Arthur’s papers collected in the volume Increasing Returns and Path Dependence in the Economy, especially his preface and my foreword for longer comments on Arthur’s work in historic perspective. Click here to see the foreword.) Hence, Krugman’s whole attack is directed at a statement made neither by Arthur nor by Cassidy. Krugman has not read Cassidy’s piece with any care nor has he bothered to review what Arthur has in fact said.

    What Cassidy in fact did in his article was to trace a line of influence between one of Arthur’s early articles and the current claims of the Department of Justice against Microsoft. It appears that Cassidy based his article on several interviews, not just one.

    The point that Arthur has emphasized and which is influential in the current debates about antitrust policy is the dynamic implication of increasing returns. It is the concept of pathdependence, that small events, whether random or the result of corporate strategic choice, may have large consequences because of increasing returns of various kinds. Initial small advantages become magnified, for example, by creating a large installed base, and direct the future, possibly in an inefficient direction. Techniques of production may be locked in at an early stage. Similar considerations apply to regional development and learning.”””

  25. Gravatar of Doug M Doug M
    22. April 2013 at 13:07

    “Surely you can accept that one huge difference between the two, is that government has limited powers to ameliorate real shocks. But the central bank has essentially infinite power to set pretty much any nominal level that it wants. Don’t you find that makes it valuable to analyze the two kinds of shocks separately?”

    Not at all.

    If my costs increase and yet I am not able to increase my prices, I will realize smaller profits. Applying more pressure, I may choose to lay off employees to maintain profitability. And at some threshold, I will be forced out of business. It doesn’t matter if my costs are rising because of “supply” or “demand” factors.

    Unless all prices move uniformly a nominal shock is a real shock. If you believe that some prices are sticky (usually wages), then all prices do not move uniformly. Hence, all shocks are real shocks.

    Should a shock hit the economy, what is the policy response? Increase the money supply until the real wage falls enough. Does it matter of the employment was a result of a real shock or a nominal shock?

  26. Gravatar of marcus nunes marcus nunes
    22. April 2013 at 13:18

    @Ashok
    You have my e-mail. Let´s move the conversatio there. We´re clogging-up this comment thread (like some usually do!)
    joaomarcus.marinhonunes@gmail.com

  27. Gravatar of ChacoKevy ChacoKevy
    22. April 2013 at 14:03

    re Boston: I don’t know if the decision to close down was a good one or not, but I do have to acknowledge that authorities never have a “right” choice. I know the weather is different from terrorism, but blizzards present a classic “damned if you do, damned if you don’t” scenario.
    Here is Rendell calling America a nation of wusses for cancelling a football game: http://articles.businessinsider.com/2010-12-27/sports/29958509_1_wusses-eagles-game-radio-station

    And here is then-candidate Rahm Emmanuel going after Daley for failing to close roads, mainly LSD, soon enough! http://chicago.cbslocal.com/2011/02/03/emanuel-backs-off-criticism-of-blizzard-response/

    At any rate, welcome back.

  28. Gravatar of Policy Wank Policy Wank
    22. April 2013 at 14:51

    The decision to shutdown any areas besides one neighborhood in Watertown was a massive overreaction. At the time of the shutdown one suspect was dead, while the other suspect was out of pocket, but very likely in the Watertown area. I guess they decided to shutdown because they lost him around 1 AM early Friday morning, so there was a small chance that he could have gotten beyond Watertown by start of business on Friday and he was still armed and dangerous. So they wanted to prevent any chance of him encountering anyone in public, which could have resulted in a few more deaths. I think it boiled down to extreme risk aversion.

  29. Gravatar of ssumner ssumner
    22. April 2013 at 15:07

    Marcus, Does Miles have a post where he explains his views?

    Steve, Then why did they lift the curfew before the suspect was captured? Did he somehow stop being dangerous?

    Ashok, You said;

    “It’s a show of strength, to the country and world. I think.”

    Seems like the exact opposite to me.

    I’m not expert on the models you cite, but as a general rule I’m very skeptical of claims that a research agenda in macroeconomics will eventually pan out. They either pan out immediately, or not at all.

    Integral, Yes, I thought about discussing why P is an absolutely horrible ONI, but given recent events in the eurozone I decided that it wasn’t even necessary.

  30. Gravatar of Steve Steve
    22. April 2013 at 15:56

    “Then why did they lift the curfew before the suspect was captured? Did he somehow stop being dangerous?”

    Classic bureaucratic over preparation, then under preparation. Kind of like Hurricanes Irene and Sandy. The cost benefit calculation always fixates on the most recent event to the exclusion of all else.

    There are lots of questions, like if he was running on foot and bleeding, how come a dog couldn’t follow his scent? Did they even try releasing the dogs? How come a police cruiser never went down the street and spotted blood on the boat? Why was the search perimeter so narrow even though the lockdown was so large — were they more concerned about police safety than public safety?

  31. Gravatar of Ashok Rao Ashok Rao
    22. April 2013 at 17:23

    Scott,

    Yeah, I agree. Overreaction isn’t a sign of strength. Just trying to understand where they might be coming from.

    Re: models – don’t you think this is to do with the culture that emphasizes clear and pure math, though? Isn’t there a clear, if slow, trend towards more computational methods?

    Anyway, if you are interested, I talk about the models at some abstraction on my blog, but the paper I linked to Marcus (http://www.santafe.edu/media/workingpapers/13-04-012.pdf) by W. Brian Arthur does a much, much better job than I could.

    Another more specific overview here: http://dare.uva.nl/document/475760

  32. Gravatar of Benjamin Cole Benjamin Cole
    22. April 2013 at 18:38

    I agree with this post 100 percent.

    On Boston: The difference, in both media and government response, to the Texas explosion and the Boston explosion is remarkable.

    More people died in Texas. They have quickly become invisible.

    While I am on the topic, about 30,000 Americans every year die in auto accidents.

    But we fear terrorism.

    Imagine if terrorists killed 10,000 a year. As Scott Sumner suggests, we would become police state.

  33. Gravatar of TravisV TravisV
    22. April 2013 at 23:03

    Miles Kimball writes:

    “An Economist’s Mea Culpa”

    http://blog.supplysideliberal.com/post/48606085691/an-economists-mea-culpa-i-relied-on-reinhart-and

    I thought he was going to admit that he mischaracterized NGDPLT! Oh well…..

  34. Gravatar of Cameron Cameron
    22. April 2013 at 23:19

    Lars Svensson is leaving the Riksbank.

    http://www.bloomberg.com/news/2013-04-22/bernanke-peer-quits-in-sweden-as-inflation-targeting-tested.html

  35. Gravatar of Blue Aurora Blue Aurora
    23. April 2013 at 02:31

    Professor Sumner, as I have said before via e-mail (and I am probably echoed by many other people’s sentiments), you and your fellow travelers in the Market Monetarist movement have contributed much to the discourse on the econoblogosphere. Your engagement with Noah Smith and the other critics of nominal income targeting is certainly a good example of this.

    However, has anyone in the Market Monetarist movement considered going beyond their specialty in monetary economics? Like say, into financial economics, or microeconomics, or international economics, or something else?

    P. S. You still need to respond to my last e-mail, sir!

  36. Gravatar of J.V. Dubois J.V. Dubois
    23. April 2013 at 02:58

    Very good post, nothing to add there. Had that “stop confusing nominal and real” fealing when reading Cochrane too. Plus I am really starting to feel confused when somebody speaks about “interest rates”. Do they mean real interest rates – but then why using the world “real” somewhere else? Or is it nominal interest rate so that they see it as some kind of ONI?

    PS: For Boston, you are exactly right. If anything it seems that USA was made more vulnerable, not stronger during last few years. In 2001 you terrorists had to hijack planes, and attack important symbols to instill fear. Now two boys with two pot bombs can actually make most powerful country in the world to hold its breath making hundreds of thousands of people being locked up at home. There is a very good article at Polish Gazeta Wyborcza about this: http://wyborcza.pl/1,76842,13782849,Boston__czyli_terror_mediow.html

  37. Gravatar of ssumner ssumner
    23. April 2013 at 05:02

    Ashok, That abstract doesn’t look very promising. I believe that “equilibrium” is a very useful concept in macro.

    Thanks Cameron.

    Blue, I already post on a number of other topics, such as taxes and China.

    JV. Good point.

  38. Gravatar of Ashok Rao Ashok Rao
    23. April 2013 at 05:19

    Scott,

    The whole idea behind the article, though, is that there might be a lot more than equilibrium. They’re not saying it doesn’t exist, or that it isn’t very useful, just that equilibrium is a special case of non-equilibrium, and there are certain processes (the evolution of political systems, for one) that a “perpetual motion” type framework might help more.

    Noah Smith also has a great post on equilibria: http://noahpinionblog.blogspot.in/2013/04/what-is-economic-equilibrium.html

  39. Gravatar of sw sw
    23. April 2013 at 06:11

    Overreaction in Boston? Maybe, maybe not. But some of the arguments along the lines of “your more likely to be struck by lightening than killed by a terrorist” strike me as really off the mark.
    Consider two scenarios. One of your neighbors stumbles and falls down the stairs, breaking his neck and killing himself. Tragic, of course. Maybe some people will be slightly more careful on the stairs for a bit. But otherwise, there really are not any public safety implications.
    The next night, a person in your neighborhood is viciously murdered and the murder is not caught. Are people suppose to take the same attitude toward this case as they did the tragic accident? Surely not.
    A human intent on harm and on the loose has qualitatively different implications than accidents or natural disasters, however tragic they may be. If that human is joined in his effort by other equally intent on harm, your enter yet another realm. If these individuals are supported by a state apparatus the implications are even more grave.
    Maybe Boston overreacted, but I’m tired of hearing that these incidents should be treated like daily car accidents.

  40. Gravatar of Patrick R. Sullivan Patrick R. Sullivan
    23. April 2013 at 06:21

    Ashok writes;

    ‘The point that Arthur has emphasized and which is influential in the current debates about antitrust policy is the dynamic implication of increasing returns. It is the concept of pathdependence, that small events, whether random or the result of corporate strategic choice, may have large consequences because of increasing returns of various kinds. Initial small advantages become magnified, for example, by creating a large installed base, and direct the future, possibly in an inefficient direction. Techniques of production may be locked in at an early stage. Similar considerations apply to regional development and learning.”””’

    I can’t, in good conscience, allow the Santa Fe Institute to continue to corrupt impressionable youth. Brian Arthur’s theory of path dependence has been destroyed many, many times. Stake after stake has been driven through its heart, yet QWERTYcula keeps coming back.

    Krugman was ‘pissed’ at Arthur for good reason (but wasn’t candid in saying why). Krugman had spent a couple of years at Stanford in the nineties where he became acquainted with the theory from Arthur and (the much more prestigiously credentialed) Paul David. He bought into it…hook, line, sinker…writing an entire chapter in ‘Peddling Prosperity’ about it.

    Then it was appointed out to him that there was absolutely no evidence supporting it. None, nada, zilch. Paul had egg all over his face. That’s why he was ‘pissed’. He’d made the mistake of taking Brian Arthur seriously. Don’t let it happen to you.

  41. Gravatar of dtoh dtoh
    23. April 2013 at 06:40

    “Did they really close down Boston when I was gone? If so, why?”

    Have to give the police and umpteen other agencies a chance to run around and play with all of their expensive toys.

  42. Gravatar of Patrick R. Sullivan Patrick R. Sullivan
    23. April 2013 at 07:19

    Btw, in the Brian Arthur paper Ashok links to we learn that an economy never reaches equilibrium, it keeps getting sidetracked and has to regroup and move toward it again.

    Which is what Schumpeter said in the 1940s in Capitalism, Socialism and Democracy.

    Brian Arthur; hemlocking is too good for him!

  43. Gravatar of Ashok Rao Ashok Rao
    23. April 2013 at 08:04

    Patrick,

    Does Kenneth Arrow have “egg all over his face” as well?

    Look, I don’t know enough to comment one way or another, but I can say that Eric Beinhocker’s Origin of Wealth (a similar argument) is very interesting indeed.

  44. Gravatar of Patrick R. Sullivan Patrick R. Sullivan
    23. April 2013 at 08:43

    Arrow seems to have been being generous to a colleague, but I don’t know of any work of his that bears on (or even shows a superficial knowledge of) the theory of path dependence.

    I do know that Hal Varian had egg all over his face from his treatment of the issue, along with Carl Shapiro, in their ‘Information Rules’. But, at least Varian was honest enough to debate me over it.

  45. Gravatar of Ashok Rao Ashok Rao
    23. April 2013 at 08:52

    Patrick,

    Arthur clearly knows about Schumpeter and, if you read the paper, gave him due credit:
    “”The other driver of disruption is technological change. About a hundred years ago, Schumpeter (1912) famously pointed out that there is “a source of energy within the economic system which would of itself disrupt any equilibrium that might be attained.” That source was “new combinations of productive means.” (Nowadays we would say new combinations of technology.) Economics does not deny this, but incorporates it by allowing that from time to time its equilibria must adjust to such outside changes.””

    As well as to Hayek and Smith (particularly Smith).

    The difference, which he clearly maintains today, is that computation can actually help people study this unlike, you know, the 1940s. The start of the paper asserts that this is nothing new.

  46. Gravatar of Patrick R. Sullivan Patrick R. Sullivan
    23. April 2013 at 09:07

    If you want to learn more, I suggest;

    http://papers.ssrn.com/sol3/papers.cfm?abstract_id=1698486

    ‘The Troubled Path of the Lock-In Movement

    ‘Paul David (1985), Brian Arthur (1989) and others introduced a ‘new economics’ of increasing returns, alleging problems of path dependence and lock-in. These conditions were claimed to constitute market failure and were soon featured in antitrust actions, most famously in Microsoft. We (1990, 1994) challenged the empirical support for these theories and their real-world applicability. Subsequently, David and others have responded, arguing that lock-in theories require no empirical support, market failures were never an important feature of their writings, and the empirical evidence that had been forward was never meant to be taken literally. Nevertheless, David and others claim that their theories have policy significance. Indeed, lock-in claims continue to appear as a basis for antitrust action. We now respond to the response of David and others, review new developments in this literature, and consider antitrust implications in light of the deficiencies in lock-in theories and related empirical work.’

    Just don’t let Brian Arthur catch you reading it!

  47. Gravatar of Patrick R. Sullivan Patrick R. Sullivan
    23. April 2013 at 09:24

    Well, Arthur says;

    ‘But this technology force is more disruptive than Schumpeter allowed.’

    From 1908! Then he goes into a tiresome explanation;

    ‘Novel technologies call forth further novel technologies: when computers arrive, they call forth or “demand” the further technologies of data storage, computer languages, computational algorithms, and solid-state switching devices. And novel technologies make possible other novel technologies: when the vacuum tube arrives, it makes possible or “supplies” the further technologies of radio transmission and receiving, broadcasting, relay circuits, early computation, and radar. And these novel technologies in turn demand and supply yet further technologies.’

    And on and on. Which Schumpeter (1942) pretty much did describe. Arthur is re-inventing the wheel.

  48. Gravatar of Patrick R. Sullivan Patrick R. Sullivan
    23. April 2013 at 09:33

    And Arthur clearly doesn’t know his Hayek, since he writes;

    ‘Complexity economics, by contrast, teaches us that the economy is permanently open to response and that every part of it is open to new behavior””to being exploited for gain, or to abrupt changes in structure.’

    That hardly would be news to the author of ‘The Use of Knowledge in Society’. Who would then understand the folly of concluding;

    ‘A complexity outlook would recommend putting carefully thought out controls in place, much as authorities put sensible building codes in place in seismic regions. But just as important, it would bring a shift in attitude in the direction of realism. The economy does not consist of a set of behaviors that have no motivation to change and collectively cause optimality; the economy is a web of incentives that always induce further behavior, invite further strategies, provide collectively “reasonable” outcomes along the way, and ever cause the system to change.’

    Which is exactly the reason Hayek gave for believing that ‘carefully thought out controls’ would not only not work, but would be further disruptive.

  49. Gravatar of Brian Donohue Brian Donohue
    23. April 2013 at 12:21

    Scott, This is a superb post. Thank you.

  50. Gravatar of Patrick R. Sullivan Patrick R. Sullivan
    23. April 2013 at 13:18

    Steve Margolis, co-author of the above linked to, ‘The Troubled Path of the Lock-In Movement’, finds as much amusement in Brian Arthur’s ‘new’ paper, as I do. He’s e-mailed me (and given me permission to quote him):

    ‘Hey, with all this stuff about the economy being living, messy, etc, maybe he’s [Brian Arthur] on his way to becoming an Austrian.’

  51. Gravatar of Jeff Jeff
    24. April 2013 at 04:32

    About the overreaction in Boston: You may recall the DC Sniper stories back in 2002. For several weeks, the police had no idea who they were looking for. That didn’t stop them from setting up roadblocks on the Beltway that kept commuters in traffic overnight. That, and the Boston shutdown, are of a piece with the TSA: We have to do something, and this is something!

    You misunderstand the purpose of DSGE models. They are a device for determining who gets tenure, nothing more. The idea that these models have improved our understanding of the economy after 40 years is laughable. Are today’s policies (say, the Euro) better than they were in the 1960’s? Is the welfare state bigger or smaller? Are we more or less free? Is unemployment more or less of a problem? Are you kidding?

    Now I’m being unfair blaming all this on DSGE modeling. But how much progress might have been made if so many talented people had not been wasting time on this crap? What if there were some incentives in the profession to reward endeavors that actually led to better policies?

  52. Gravatar of ssumner ssumner
    24. April 2013 at 05:53

    Ashok, OK, but the proof is in the pudding. What do they explain better than MMs?

    Thanks Brian.

    Jeff, But at least they didn’t shut down all of DC!

  53. Gravatar of Ashok Rao Ashok Rao
    24. April 2013 at 05:59

    I’m not sure what policy relevance there is, yet, but models from Epstein and Axtell at Brookings like Sugarscape do a very interesting job of the emergence of inequality and wealth in society based on simple changes in trading rules and patterns, as well as natural endowments. This was a long time ago (computationally speaking), a lot more is possible today.

    But from Mark Buchanan, here’s a better example of where these might be appropriate:
    “Consider, for example, the assertion of some prominent economists, such as Stanford University’s John Taylor, that the low-interest-rate policies of the Federal Reserve were to blame for the housing bubble. Some dynamic stochastic general equilibrium models can be used to support this view. The agent- based model, however, suggests that interest rates weren’t the primary driver: If you keep rates at higher levels, the boom and bust do become smaller, but only marginally.”

Leave a Reply