Pu Pu Platter

I woke up today dreading having to write part 2 of my liberalism and utilitarianism post, and then decided to just blow it off.  I suppose a blog should be more like jazz improv, rather than the laborious construction of a symphony in parts.  So tonight I’ll do shorter pieces on the following 4 questions:

1.  Are macroeconomists just a bunch of astrologers?

2.  Are Democrats just a bunch of socialists?

3.  Do soaps promote liberal values?

4.  How many Tyler Cowens are there?

1.  Are macroeconomists just a bunch of astrologers?

In a recent bloggingheads.tv episode, Mark Thoma made an interesting observation about modern macroeconometrics:

We have one set of [U.S.] data.  . . . We know what that data says.  . . . It’s hard to build a model that doesn’t fit.

That got me thinking about my skepticism regarding macroeconomists who try to build structural models of the economy.  Let me say right up front that I don’t know a lot about these models, and it’s really dangerous to summarily dismiss a whole line of research that one isn’t very well-informed about, but . . . you know what’s coming next.

It seems to me that structural models of the economy probably have some ability to extrapolate into the future, but don’t seem to be able to predict the things that we’d really like them to be able to predict—sharp changes in AS or AD.  I think about that failure in terms of efficient markets theory.  Almost all the really interesting macro events that I have lived through or studied have been accompanied by dramatic swings in either stock, commodity or bond prices.  Unless these swings are predictable, and I’ve never seen any convincing evidence that they are, then I don’t see how structural models can be of much use in stabilization policy.

Whenever I read opinion pieces by almost any macroeconomist— Keynesian, monetarist, new Classical, Austrian, etc, there is almost invariably a point where alarm bells go off.  At some point the economist will make an assertion that seems to me to be in conflict with the EMH.  And after that point I have trouble taking anything they say seriously.  I keep thinking “If you’re so smart . . . ”

If you read through all my posts you will have trouble finding a single assertion that is at variance with the EMH.  In contrast, pick up almost any other economist’s take on the current crisis and you will almost invariable find at least one assertion that conflicts with the EMH.  It’s always something like “the root cause of the crisis occurred years earlier when . . .”

1.  The Fed set interest rates too low

2.  Regulators let banks make excessively risky loans (or if you’re a right winger–encouraged them to make risky loans.)

3.  Americans didn’t save enough

And so on.  In contrast, I believe that the depression was caused by events that took place in September and October, when the markets actually crashed. Which depression?  All of them—1929, 1937, 2008, etc.  And as far as I know I am the only economist who believes that all of these depressions were caused by events that occurred in those two fateful months.  (Although I know that Earl Thompson has a similar view of at least the current crisis, as do a few readers of this blog.)

When I talk to people who believe in astrology I sometimes get the mischievous urge to provoke them with questions.  OK, if one’s sign determines one’s personality, wouldn’t that be easy to test?  In any large group of type A personalities (say the 535 Congressman) shouldn’t a disproportionate number have been born in certain months?  I see a flicker of uncertainty in their eye, before I invariably get a reply along the lines of “well, it’s more complicated than that.  There are a number of factors.”  OK, fair enough.  But then why bother asking people “what’s your sign?”

When I talk to other economists about their theory of the 1929 crash, the 1987 crash, the 2000 crash, the 2008 crash, etc, I also try to provoke them with questions about efficient markets.  If Fed policy was obviously far off base, and if the policy was inevitably going to lead to a crash, why did so many smart investors ignore those signs?  Why couldn’t anyone using your model have gotten rich?  “Well, it’s hard to predict the exact peak.”  But why do you need to predict the peak?  Asset markets are incredibly volatile, why not simply buy when markets are more than 20% undervalued and sell when they are more than 20% overvalued?

I have strongly argued that only the first, very mild part of the crisis was attributable to the sub-prime fiasco.  But many commenters argue with me, asserting the entire crash, even the past year, is a relapse from an easy money-induced speculative orgy into real estate.  I can’t imagine why anyone who believed that wouldn’t be rich by now.  I know the Feds cracked down on shorting bank stocks late last year, but surely there must have been other ways to short the market.  I can’t believe those regulations would have thwarted a determined bear.

My criticism applies to everyone.  It applies to the Ivy league-types who arrogantly think their computer models can predict better than asset markets.  But it also applies to Austrian economists who have contempt for those mathematical models of the economy.  It even applies to people who should know better, new classical economists like Robert Lucas.

How could a new classical economist think that they know better than asset markets?  I don’t know, perhaps they are not aware that their policy views conflict with the EMH.  A recent opinion piece by Robert Lucas started off promisingly, seeming to get all the big issue right.  Unlike like many right wingers, Lucas realized that we faced a severe problem of falling AD.  Unlike many left wingers Lucas realized that liquidity traps don’t preclude aggressive monetary stimulus.  I was all primed for Lucas to agree with my views, when he suddenly announced that Bernanke was doing the right thing.  This was a big disappointment to me (as Lucas was my advisor at Chicago.)

Why did Lucas think policy was on target?  He ignored all the information packed into asset prices.  He ignored the clear signs that inflation and NGDP are expected to come in far below target.  Instead he simply looked at what Bernanke had done (a big increase in the monetary base) and assumed that in time this policy would work its magic.  We will get a recovery someday, perhaps soon.  But we have known at least since last October that asset markets view the actions of the Fed as being woefully inadequate (for reasons that are not hard to figure out.)  So even the vanishingly small group of economists who share my view of the potency of monetary policy, the insufficiency of AD in late 2008, and the rationality of expectations, even they (he?) departed from my policy critique.

When I point out to someone that their explanation of the crash has obvious investment implications, and that if true they should be rich, they tell me that it’s much more complicated than I make it out to be.  Believe me; I do believe the root causes are extremely complicated—far too complicated to be explained by highly technical structural models developed at MIT, or highly subtle non-technical models from the Austrian school.  So what’s my explanation?  Don’t I have a model?  Well, there are models and there are models.  I am criticizing models that attempt to find root causes for the recent crash, which should have allowed investors to anticipate a crash.  I deny there are any models that can predict sharp breaks in AD, in NGDP.

On the other hand we have a very good model of what happens when NGDP does fall sharply.  That model was developed by David Hume, and came through this crisis (like the crises of 1929 and 1937) with flying colors.  That model predicts that if NGDP falls sharply, RGDP will also fall sharply.  As Friedman once said, in 200 years all we’ve done is go one derivative beyond Hume.

It’s hard to overcome our deep-seated instinct that it ought to be possible to look at these crises, which look like a sort of morality play, and discern useful patterns.  One commenter argued that bursting bubbles cause recessions, not tight money (as I argue.)  When I pointed out that the 1987 crash was followed by a nearly three year boom, he continually insisted that it was followed by a recession, not a boom.  We debated this issue back and forth.  He must have been a Taurus.

2.  Are Democrats just a bunch of socialists?

I don’t have much interest in talk radio anymore, but I gather there has been some recent controversy about people like Rush Limbaugh calling liberal Democrats “socialists.”  Democrats seem offended by this charge.

A recent post by Schrivener.net showed some interesting poll results.  It seems that Republicans favor capitalism over socialism by an 11-1 margin, whereas Dems favor capitalism by only a 39-30 margin.

I should say right up front that such polls are not very meaningful, as terms like ‘capitalism’ and ‘socialism’ mean very different things to different people.  Even so, are you as surprised as I am that 3 million Republicans say they favor socialism?  (There was obviously a “neither” choice offered as well.)  Regarding the 30% of Dems who favor socialism, I suppose it might be the case that they are simply referring to a social-welfare state like Denmark (and perhaps they didn’t read my blog post pointing out Denmark is the most free market economy on Earth.)  I suppose that would be the explanation of liberal pundits offended by conservatives who charge that the Democratic Party is infested with socialists.

Anyone who reads the blogosphere, however, knows how liberal Democrats regard terms like ‘deregulation’ and ‘privatization.’  I think it’s fair to say that their goal is not to make America the most free market economy on Earth.  On the other hand, they clearly don’t want a Soviet-style economy either.  So what do they want?  The only constant that I have been able to ascertain is that they want more government in areas where there are problems.  So after Great Depression we added the FDIC, and the SEC, and many other regulatory agencies.  After the S&L debacle of the 1980s, regulation of the banking industry was tightened.  After Enron it was tightened again.  You might assume that the current banking crisis, the worst ever in terms of the way banks actually behaved (although not as bad as the 1930s in terms of bank failures) has led liberals to re-evaluate their faith in regulatory fixes.  You would be wrong.  Once again we are told that we need much more regulation.

I’m obviously not the first to notice that this procedural method, regulate more after each market failure, even if it is actually a regulatory failure; is a one-way ticket to socialism.  Fortunately things aren’t nearly as bad as they seem (although they are plenty bad right now.)  There are two forces tending to check the tendency toward ever greater regulation; liberal Democratic economists, and conservative Republican voters.

Although the common sense perspective of most liberals is that socialism offers answers to many of our problems, liberal economists (sometimes) know better.  Thus in the 1970s they did not respond to excessively high air fares by urging tighter price controls, but rather they convinced President Carter to abolish the CAB.  They did not respond to the failures of AFDC by supporting more welfare spending, but rather convinced President Clinton to “end welfare as we know it.”  Yes, I know things are going the other way right now, but just wait ten years.  (The only op-ed article I ever published predicted (in early 1993) that Clinton would soon run into trouble, and then turn to the right.)

The other group restraining liberal Democrats is conservative Republican voters.  Is it possible that (contra J.S. Mill) conservatives are actually smarter than liberals?  I should probably stay away from a direct answer, as I would offend just about everyone.  Let me put it this way.  From my perspective (a right wing liberal) conservative voters often seem right about economics, but not always for the right reason.  I assume they have the same sort of commonsense view of the world that many Democrats have, but a different value system.  In particular, they put less emphasis on purely utilitarian considerations, and more on religion, tradition, and getting one’s just deserts.  Of course both parties contain a whole lot of people motivated by self-interest—low income Democrats or public employees who favor bigger government, and Republicans who resent paying taxes.  But self-interest doesn’t explain why people vote.

Is there any connection between this essay and the previous one?  Maybe that people have trouble seeing things clearly when the issues are highly emotional.  Even uninformed non-economists have incredibly strong views on the causes of economic crises like the current one.  It is hard for people to look dispassionately at these sorts of traumatic events.  Politics seems the same.  People think those on the other side are either fools or knaves (or both.)  Many people (on both sides) are fools or knaves, but as long as you look at things that way you won’t really understand politics.

3.  Do soaps promote liberal values?

The Economist magazine says yes:

The soaps blossomed under Brazil’s military regime of 1964-85.  . . . Their scriptwriters and directors, many of whom were on the left, saw them as a tool with which to reach the masses. Their plots often tilt in a progressive direction: AIDS is discussed, condoms are promoted and social mobility exemplified.

How much impact do the soaps have on real life? As recounted in papers from the Inter-American Development Bank, researchers tracked Globo’s expansion across the country and compared this to data on fertility and divorce.*

The results are most striking for the total fertility rate, which dropped from 6.3 children per woman in 1960 to 2.3 in 2000, despite contraception being officially discouraged for some of that time.  . . .  Controlling for other factors, the arrival of Globo was associated with a decline of 0.6 percentage points in the probability of a woman giving birth in a given year.  . . .

The effect on divorce was smaller, but noticeable. The researchers found that between 1975, when divorce was first mooted, and 1984 about one in five of the main characters in Globo soaps were divorced or separated, a higher percentage than in the real Brazil. These break-ups were not just a result of machismo: from the mid-1960s to the mid-1980s about 30% of female lead characters in novelas were unfaithful to their partners. The researchers find that the arrival of Globo in an area was associated with a rise of 0.1-0.2 percentage points in the share of women aged 15-49 who were divorced or separated. The authors reckon that watching “empowered” women having fun in Rio made other women (a few of them anyway) more independent.

Other research shows that divorce and lower fertility are linked to less domestic violence. So the influence of soaps may be far more positive than critics of their vapidity claim. If Globo could now come up with a seductive novela about tax reform its transformation of Brazil would be complete.

That last line pretty much summarizes my earlier post on political art; cultural values—yes, tax reform . . . not so much.  The “vapidity claim” reminds me of G.K. Chesterton’s “A Defense of Penny Dreadfuls”:

One of the strangest examples of the degree to which ordinary life is undervalued is the example of popular literature, the vast mass of which we contentedly describe as vulgar.  The boy’s novelette may be ignorant in a literary sense, which is only like saying that a modern novel is ignorant in the chemical sense, or the economic sense, or the astronomical sense; but it is not vulgar intrinsically—it is the actual centre of a million flaming imaginations.

By the way, Chesterton and I could not be farther apart politically, but he was ruthlessly effective at exposing the prejudices of intellectuals.  Although I despise soaps, I see no reason to deny their obvious benefits to society.  You might ask whether this study is to be believed, after all, correlation doesn’t prove causation.  But this type of study is the gold standard of social science research.  The soaps were rolled out across Brazil one area at a time.  So one couldn’t argue the correlations represented more liberal women choosing to watch soaps.

What’s the relationship with the previous two essays?  Stubbornness.  I saw a similar study that looked at the effect of TV on school performance in America.  TV was also rolled out city by city across the U.S., and thus any sociological effects should be easy to identify in the data.  In fact, TV does not hurt school performance.  But when I tell this to people they simply don’t believe it.  Perhaps they rely on crude correlations they have observed during their lives (kids who don’t like reading presumably watch more TV.)  Perhaps introspection.  However I doubt that even introspection can explain this.  I find that while people feel the media are very powerful, they see themselves as somehow immune to its mysterious influence.  It is always “those other people” who are swayed by the media.

So people put great faith in their own gut instincts, even if based on methodologically worthless techniques, but summarily discount a study that identifies exogenous shocks more effectively than just about any other social science research.

4.  How many Tyler Cowens are there?

For those who don’t know, Tyler and Alex Tabarrok run the number two economics blog, MarginalRevolution.com.  Tyler has far more posts than I do.  Yes, his are much shorter, but look at the research behind each one.  You might think the title question here is metaphorical, referring to Tyler’s expertise in what seems like virtually every area of intellectual life, not to mention all sorts of pop culture trivia.  But I meant the question literally.  Does he have a little helper like George Will has?  How about a quantum machine, allow him 96 hours a day reading in 4 parallel universes?  Clones?  Your guess is as good as mine.

My blog takes up about 6 hours a day.  As a result I have cut out large parts of my former life.  I am a big film buff, and yet have only seen one new film this year (Kore-eda’s newest.)  Recently I have become very depressed as I read Tyler’s blog each day.  Almost everyday he seems to produce a list of new books he has read.  The last novel I read was finished a few days before I started this blog—the 900 page tome just stares at me reproachfully from the nightstand.  I feel as though my former life was acquiring knowledge, and now I am just disseminating old ideas.  I have resorted to strip-mining old unpublished papers, while I do no new research.  I am getting dumber.

When I started this blog I had an ideal reader in mind (as I suppose all authors do.)  I knew that it was going to be impossible to convince macroeconomists of my extremely counter-intuitive views of the crisis.  Nobody who has expertise in macro can be impartial to an event so traumatic.  By now views have hardened.  Even the uninformed have strong opinions.  So I needed someone outside of economics, someone with an open mind.

At the same time I knew that I needed a reader that was bright enough to understand very subtle, counter-intuitive economic arguments.  Thus I needed someone who had at least an amateur’s interest in macro, who was also a very skilled economist.  I also needed someone who wasn’t overly impressed with a lot of technical mumbo-jumbo that doesn’t really mean anything.  Someone who thought short (or not so short) blog posts could convey interesting economic ideas.

In the first three or four weeks (when almost no one was reading my blog) I deliberately tried to make my writing more appealing to this ideal reader.  So it looks like my well thought out strategy paid off.  At least I thought so until today, when I listen to the last part of an old bloggingheads.tv that I hadn’t had time to finish earlier.  In a chapter entitled Has fame made Tyler boring?, Tyler was responding to Robin Hanson’s observation that he had become more conventional, when he suddenly announced:

“I would rather read weirdos.”

And I thought it was my brilliant arguments.  Seriously, I do think that it is easier to have unconventional views when you are isolated.  When I talked to Mark Thoma I probably (subconsciously) thought it would be impolite to just throw out a bunch of wild generalizations (“all mainstream macroeconomics is bankrupt”—that sort of thing.)  If I was giving a seminar at the Fed Board of Governors tomorrow, I would tone down the rhetoric a bit.  At the same time I find counter-intuitive arguments very appealing—and I suspect this is true of Tyler as well.

I think one reason that Tyler may now seem more conventional is that age brings wisdom, which brings a more detached view of life.  You start to see things from many different angles.  You see life in less Manichaen terms.  (To understand all is to forgive all.)

So how does this relate to the earlier essays?  Tyler is surely one of the most open-minded intellectuals around today.  (And even that is probably an understatement.)  He would not simply dismiss the Brazilian soap study because it conflicted with his pre-existing prejudices.

Most intellectuals have very big egos, and that can inhibit forming a clear-eyed view of the world.  I’m no exception, far too sure I that am right than I should be from any sort of dispassionate perspective.  I understand that even in the unlikely event that I have stumbled onto the “truth,” I won’t be able to “persuade my peers” with this blog.  Instead, I would refer back to what Keynes said about Gesell, who he initially viewed as a monetary crank:

“their significance only became apparent after I had reached my own conclusions in my own way.”

Update (4/27/09):  In part one I should have mentioned a paper that I did with Aaron Jackson entitled “Velocity Futures Markets:  Can They Eliminate the Need for Structural Models?” (Economic Inquiry, 2006.)  Unlike me, Aaron does understand structural models, so I wasn’t just shooting from the hip.


Tags:

 
 
 

108 Responses to “Pu Pu Platter”

  1. Gravatar of Devin Finbarr Devin Finbarr
    26. April 2009 at 22:10

    At some point the economist will make an assertion that seems to me to be in conflict with the EMH. And after that point I have trouble taking anything they say seriously. I keep thinking “If you’re so smart . . . “

    The major reason these crashes are so hard to predict is that they depend on government actions. Imagine the Fed announced a completely Laissez Faire policy: no new money would be ever be printed, the FDIC would not be bailed out, no banks would be bailed out, no money market funds would be bailed out, no one was too big to fail, etc. If this happened, the correct action would be obvious – hoard dollar bills and gold in safety deposit boxes. Liquidate everything.

    But in a world where Bear Stearns and AIG get bailed out, while Lehman is allowed to go bankrupt, investing becomes entirely dependent on government action.

    Consider the choice of whether to put money in a money market fund. One person might argue that only suckers would put money in such a fund. Some of the funds have assets with an average maturity of 30 years. In a run, you might find that your dollar is gone and replaced with an IOU for 30 years from now. But another person argues that the Fed tracks money markets as part of the money supply, and even if they are not FDIC insured, the Fed would never let them fail. So you might as well put your money in them, and take the extra interest. And it turns out, the money markets have indeed been bailed out. The prudent investor lost money, to the investor who had the correct political intuition.

    Or consider the stock market investor. In a world of zero monetary growth or an NGDP target of 0%, when a stocks trade at 3% dividend yields, it would be crazy to buy them. You’d be much better off investing in corporate bonds. But when the money supply dilutes at ~8% a year, while stocks dilute at only 2% a year, and CD’s only pay 3% interest, then stocks could potentially make a very good investment, despite the low dividend yields. But again, investors in the stock market have no way of knowing when the money supply is suddenly going to contract.

    It should be possible to price stocks, bonds, and housing based on yield. But in a world of unpredictable federal bailouts, and 30 year booms propped up by easy money, “too big to fail” bailouts, and the Greenspan put, everyone becomes a speculator.

  2. Gravatar of The Ambrosini Critique » Blog Archive » Blog Post of the Year The Ambrosini Critique » Blog Archive » Blog Post of the Year
    26. April 2009 at 23:18

    [...] won’t be a contest this year. This post [...]

  3. Gravatar of O clube de convergência inferior do debate « De Gustibus Non Est Disputandum O clube de convergência inferior do debate « De Gustibus Non Est Disputandum
    27. April 2009 at 00:20

    [...] mesmo dia, ou quase, Sumners fala da relevância dos macroeconomistas e também discute rótulos e afins. Sobre a blogosfera, veja o que ele diz: When I started this [...]

  4. Gravatar of Nick Rowe Nick Rowe
    27. April 2009 at 02:26

    “I feel as though my former life was acquiring knowledge, and now I am just disseminating old ideas. I have resorted to strip-mining old unpublished papers, while I do no new research. I am getting dumber.”

    I disagree on that point. Even if you are disseminating (your) old ideas, you are applying them to the current problems, and arguing them out with a very varied group of people, with very different views of the world, some of whom know a lot more about some things than you know (just as you know a lot more about other things than they know). A group of people you would be unlikely to meet in a departmental seminar. Your argument with them is public, recorded, and anyone can join in if they think you are wrong, or have missed something.

    How is that not (part of) research? How does that make you dumber?

    But yes, I know what you mean about Tyler. I don’t know how he does it. I can only blog the amount I do because I am on sabbatical this year, and do little else. It worries me that I do little else. But I have never felt as intellectually engaged in economics as I am now since I left grad skool. Things will have to sort themselves out, somehow.

    On the EMH: I expect someone could argue that X caused the current recession, but he wasn’t watching X (or data on X were not available except with a long lag), so the causation is clear only in hindsight. Not necessarily plausibly. My own view is a multiplicative mix of yours and the mainstream view. Multiply financial fragility times monetary policy mistakes. With a healthy financial system, monetary policy mistakes don’t matter as much, and can be corrected easily. With a fragile financial system, the pole the Fed is trying to balance upright gets shorter, and will fall over quickly unless the Fed fixes its mistakes very quickly.

  5. Gravatar of ssumner ssumner
    27. April 2009 at 04:37

    Devin, Good points. I actually agree with much of what you say—I think uncertainty over government policy is a huge problem for investors, and I understand your point about the “fundamentals” investor vs. the politically savvy one. Take a look at my longer answer to Nick (the part on causality), I’d be interested in whether you think it addresses your comment.

    Ambrosini, Thanks. You made my day.

    #3, Can anyone translate Portuguese? Is this good or bad?

    Nick, As soon as I saw your name I knew you’d object to my “dumber” point. I completely agree with you that the blog has sharpened my monetary economics skills. The conversation with people like you has moved me forward. Indeed even the commenters that I get most frustrated with are a big help, because they force me to defend my ideas, and that makes me think deeper into the issue. In the future more and more of my posts will be responses to those comments.

    I meant dumber in the broader sense, less well rounded (as I have less time for outside reading, films, etc.) I used to follow current events closely, now it’s sort of a big blur–I couldn’t even tell you how Obama’s doing from the perspective of the news media.

    On causality, I’d say the way you describe your view is EXACTLY my view. I am working on a causality post. Not being a philosopher, I find this issue tricky. There is no doubt that many of the factors people cite were “root causes” in some sense. I’m not blind to the fact that without the 2007 sub-prime crisis there would have been no deep recession in late 2008. I suppose this gets into the necessary/sufficient condition question. To anticipate my post, I will argue that in economics the causality question should be framed in terms of policy implications. I think there are useful policy implications for my “tight money in the autumn of 2008″ thesis, and indeed the Fed had enough information to act even back then. Regarding deeper root causes, as you put it “causation is only clear in hindsight.” To me, that’s the crucial difference. That’s why I see financial markets as the key to macroeconomics, they tell us the information set available to policymakers at each point in time.

    The decision by Mr. and Mrs. Hitler to have Adolf may have been a root cause of WWII, but there are no policy implications that flow from that fact. In contrast, markets were telling the Fed that policy was way too tight last fall, they just ignored the markets. I guess the debate is over whether (as I argue) it was not too late last fall for M-Policy to prevent the subprime fiasco from spreading into a broader decline in AD in all sectors, and in almost all countries. Maybe NGDP futures . . .

  6. Gravatar of Aaron Jackson Aaron Jackson
    27. April 2009 at 05:30

    Hey Scott,
    Just to clarify, the title of our published paper is “Velocity Futures Markets: Does the Fed Need a Structural Model?” To be fair to Scott, he referenced and earlier version of the title of our paper. And since he puts a lot of time and effort into this great blog, we can forgive him for that minor oversight, can’t we?

  7. Gravatar of Jason Jason
    27. April 2009 at 05:54

    First time commenter and former Bentley economics student, but I’ve been following your blog for a while.

    I have to disagree with your statement on getting “dumber”. In my personal experience (which is very short), you go through phases in life. You rapidly accumulate knowledge, I picture a mass of knowledge swirling inside your brain. I’m in that stage right now, and I have a lot of trouble articulating and explaining everything I’ve accumulated. I have started to be able to spit out general points, but when I am forced to dig into my brain and pull out the support, I struggle.

    This blog gives you an opportunity to learn and refine your ability to communicate your ideas and the vast amount of knowledge you’ve accumulated. It is one of life’s most important skills. No matter how much knowledge you have, if you can’t articulate it, its useless. Maybe its a bit like the business cycle, you accumulate, then you empty, then you accumulate, empty, repeat. I only hope that one day I will have a medium on which I can spew the vast collection of knowledge I have collected.

    I think the more you blog the shorter your posts will get. You will learn how to communicate your points more efficiently, and this will increase your readership (I always have time for a Tyler Cowen post, but its hard to find time at work to sit down and read one of yours sometimes). Your strategy is great right now. Your defining yourself to the world right now. It separates you from the countless other bloggers out there. Once your defined (assuming you want and can continue to blog), you’ll find yourself writing shorter posts. The blog will take up less time because your views are established and you can simply “tweak” them and present new information. You’ll be able to spend more time reading, watching films, and absorbing information.

    Best of luck and keep writing.

  8. Gravatar of Alex Golubev Alex Golubev
    27. April 2009 at 06:22

    #4 – look up the “online monoculture” study. Internet leads to much greater variety for the individual, while narrowing the scope of the collective. A positive and a not so negative. Maybe that will help you figure out what you want from the blog
    #1 – i think it’s like arguing who started a fight. I didn’t wash the dishes, you said something mean, i sucked out some leverage from the system, and you didn’t replenish it, now we don’t have milk for cereal in the morning! :) arguing what caused the crash is like arguing what caused the bubble (real estate or dotcom or tulips). i’d say self reinforcing process itself is to blame. but that process is highly important in other aspects of life (say, putting on a face mask in mexico city). Yes the scale was smaller when it was “just” a real estate slowdown, but some people go on shooting sprees after losing their job. i’d say losing your job is not THAT important either. Instead of calling Fed’s mistake a cause, i think we should call them even further blunders that made everything even worse.

  9. Gravatar of Bill Woolsey Bill Woolsey
    27. April 2009 at 06:41

    Eu falo o portugues um pouco.

    mesmo dia, ou quase, Sumners fala da relevância dos macroeconomistas e também discute rótulos e afins. Sobre a blogosfera, veja o que ele diz

    The same day, or almost, Sumners speaks of the relevance of macroeconomists and also argues labels and similar things. About the blogosphere, see what he said.

    A brazialian friend had to help me with “rotulos.”

  10. Gravatar of Bill Woolsey Bill Woolsey
    27. April 2009 at 06:47

    Scott,

    Why did you combine four different posts in one?

    To me, they all stand alone.

    You could have written four shorter ones.

    Bill

  11. Gravatar of Jason Jason
    27. April 2009 at 06:52

    Bill,

    That’s why its called “Pu Pu Platter”

  12. Gravatar of Blackadder Blackadder
    27. April 2009 at 07:15

    I don’t know if it makes you feel any better, but I suspect that Tyler doesn’t actually read all of the books that he talks about on his blog. So maybe there are only two or three Tyler Cowen’s, rather than six or seven.

  13. Gravatar of Robin Hanson Robin Hanson
    27. April 2009 at 07:43

    I’d to see other macro economists respond to your claim that they all assume big EMH violations. It seems a damning critique if true; so is it true?

  14. Gravatar of Mike Mike
    27. April 2009 at 08:28

    Great posts!

    If I think my neighbors are all spending too much on their homes, and their home prices will decrease in 3 years, what’s my arbitrage move? How can I go short someone else’s mortgage? (Especially if I don’t know where they have their mortgage.)

    Do you read any Fischer Black on macro issues?

    If memory serves, Shiller and a few others did a bunch of stuff where he tried to short the ‘irrationally exuberant’ market of the late 1990s and got hosed. That’s good old fashioned Limits to Arbitrage – the markets can stay irrational longer than you can stay solvent….

  15. Gravatar of Adam P Adam P
    27. April 2009 at 08:46

    Scott says “But we have known at least since last October that asset markets view the actions of the Fed as being woefully inadequate (for reasons that are not hard to figure out.)”

    This sounds like a statement that stock prices are currently low. If that’s what Scott meant then it’s just wrong. The long-term P/E ratio of the SP500 is just about 15 in the Shiller data, right around its long term average.

    Stock prices of a couple of years ago were extremely high but that, of course, is not a statement against the EMH in any way.

  16. Gravatar of Dan in Euroland Dan in Euroland
    27. April 2009 at 10:45

    Scott,

    Fascinating post. I too am in awe at Cowen’s reading speed and retention rate. I suspect reading speed and retention are complements for Tyler, but substitutes for the rest of us. Mere mortals that we are.

    Also, sorry to hijack the thread, but is there anyway you could post pdf links to your papers? Not all of your readers have journal access, but would probably be interested in reading the papers. I think your work could get a wider readership that way.

  17. Gravatar of johnleemk johnleemk
    27. April 2009 at 13:13

    Scott,

    Re your ideal reader, not to toot my own horn, but I feel like I am more or less your ideal reader! As long as your blog posts are, I look forward to each and every one of them — they hit the spot like few other econ blogs out there, at least for me.

    I also don’t happen to think you’re weird — but that’s probably because I’m not quite intelligent/educated enough to follow your finer points, while I tend to agree with almost everything that I can follow. One defining aspect of your blogging, I think, is that you take a relatively rare pragmatic libertarian viewpoint — most econ blogs are either on the technical side, or overtly left liberal/hard right libertarian. (Besides The Money Illusion, the only other prominent econ blog I can think of that doesn’t fall neatly into either ideological category is Marginal Revolution.)

  18. Gravatar of ssumner ssumner
    27. April 2009 at 14:10

    Thanks Aaron.

    Jason, Thanks. Very good advice. You are probably right that I will gradually move to shorter posts.

    Alex, I plan a post on causation. it’s a tough problem and seems to be the point of contention in lots of our debates. I like your analogy (job loss and shooting, sub-prime errors and Fed errors) the relative magnitudes are about right.

    Bill, Thanks for translating. I am used to doing longer posts, but I see your point. On the other hand I never would have won Ambrosini’s “Post of the year” (see above) so I don’t regret it. There is a slight connection between them.

    Blackadder, I am pretty sure he reads the fiction all the way through. Perhaps he skims some of the nonfiction.

    Robin, I’m sure other economists wouldn’t agree with my characterization, but that’s how I interpret many of their policy views. I should really do a longer article (not post) on this topic.

    I see economic disasters accompanied by sharp asset price changes, and assume that the information about the economic disaster hit the markets at the time they crashed. In that case there are no policy implications from earlier events that may have been root causes, unless one could plausibly argue that policymakers could see them next time around. But that is almost never the case (and certainly not in the subprime fiasco case.)

    The only other economist that I know of who argued that the Great Depression was caused by policy errors occurring in late 1929 was Jude Wanniski, and his argument (centering on the fight in Congress over Smoot-Hawley) wasn’t a plausible explanation of the crash of 1929. The Friedman-Schwartz view of the GD is wildly inconsistent with the EMH, as I will show if my book ever finds a publisher. Many of the famous events discussed by Friedman and Schwartz did not affect asset prices in the way their theory predicts. I don’t know of anyone else who has made that critique in any sort of comprehensive way. (These include the OMOs of 1932, the discount rate increases of 1931, the RR increases of 1936-37, etc.) And they had no explanation for the 1929 stock crash. Many I should do a post on this.

    I’ll do the rest later tonight. Thanks for all the comments.

  19. Gravatar of ssumner ssumner
    27. April 2009 at 15:20

    Mike, You could sell banks stocks short. BTW, I am not giving investment advice, as I don’t believe people can beat the market (except Warren B.)

    Shiller lost money because he was wrong in saying stocks were overvalued in 1996. I don’t know what stocks are worth, and neither does Shiller.

    Adam, As I just said to Mike, I don’t know what stocks are worth. I noticed that market expectations of inflation and NGDP growth fell sharply last October. In some other posts such as “Some graphs” (which was a few weeks back) you will find some evidence. But the hypothesis isn’t particularly controversial. The market forecasts were confirmed by the consensus private forecasts, which also showed sharply falling growth expectations. I do not favor having the Fed peg stock indices.

    BTW, I view long term P/E ratios as being of little value in investment decisions. No one knows whether the long run P/E is the right one. Ditto for long run housing values (at least in the coastal markets–maybe in Arizona and Nevada they would have been useful.)

    Dan in Euroland, I will look into PDF’s this summer. If there is something you really want to see I might be able to email it to you. My futures targeting paper at BEJ, Contributions to Macroeconomics is available online.

    Johnleemk, Thanks. You are probably also a pretty ideal reader. BTW, your final sentence provides another reason why Tyler was ideal.

  20. Gravatar of Larry Larry
    27. April 2009 at 15:44

    In the spirit of short posts, I’d love something on stagflation, that condition in which (if I have it right) inflation is kicking nominal numbers upward, while real numbers lag far behind.

  21. Gravatar of Jon Jon
    27. April 2009 at 18:05

    Scott:

    “In contrast, pick up almost any other economist’s take on the current crisis and you will almost invariable find at least one assertion that conflicts with the EMH. It’s always something like “the root cause of the crisis occurred years earlier when . . .” 1. The Fed set interest rates too low”

    How does a EMH describe the Fed? Your earlier question to the Austrians is better: why is the public fooled by the Fed’s interest-rate policy. That’s an EMH question/challenge to the Austrians.

    … but then you might wonder why the Fed is able to change the time-path of the economy by depressing the short-rate below the natural-rate at all. Why doesn’t the public just figure out?

    Obviously there must be something physical changing for real, i.e., base creation. But if that’s so… the EMH is not necessarily a rebuttal.

    So you’re down to the question: why doesn’t the economy expand in a structurally sound fashion (which is essentially what the Austrian’s suggest). Okay, that’s an EMH question.

  22. Gravatar of Adam P Adam P
    27. April 2009 at 21:24

    Well Scott, you seemed in the post to be saying you don’t think Robert Lucas really understands the situation. Your evidence, other than him not agreeing with your views, was the markets. However, one of the most basic valuation metrics suggests the market is reasonably priced, not at all low, one could certainly be reasonable to interpret that as saying the market supports Lucas’s view that Bernanke’s doing a good job and recovery will come fairly soon.

    After all, you give a critique that you say applies to everyone, you then suggest: “perhaps they are not aware that their policy views conflict with the EMH.” Your whole critique rests on it being true that all these people’s policy views conflict with the EMH but you give neither argument nor evidence to back it up. If you don’t think the P/E ratio is meaningful then fine, suggest some evidence that supports your view. However, your resonse was that you don’t know what stocks are worth which, by the way, is a statement against the EMH.

  23. Gravatar of Sören Höglund Sören Höglund
    27. April 2009 at 22:59

    As a curmudgeonly empiricist, I’ll have to pick a nit and point out that this:

    “When I talk to people who believe in astrology I sometimes get the mischievous urge to provoke them with questions. OK, if one’s sign determines one’s personality, wouldn’t that be easy to test? In any large group of type A personalities (say the 535 Congressman) shouldn’t a disproportionate number have been born in certain months?”

    is not quite the same as this:

    “When I talk to other economists about their theory of the 1929 crash, the 1987 crash, the 2000 crash, the 2008 crash, etc, I also try to provoke them with questions about efficient markets. If Fed policy was obviously far off base, and if the policy was inevitably going to lead to a crash, why did so many smart investors ignore those signs?”

    The first being hard fact, while the second contains two major assumptions: a) the majority of investors are smart, and b) smart people are not prone to irrational behavior.

    I’m willing to accept a), but not b). It’s entirely possible that members of Wall Street, when presented with views conflicting with conventional wisdom and models chose to go with the familiar. Especially if the conventional wisdom tells you everything is going to be alright, and the contrarians are telling you you’re headed for disaster.

    I’m not saying there isn’t good reason to be skeptical of grand theories, just pointing out that “why did so many smart investors ignore those signs?” is a lot less compelling and easier to counter (you could point to Taleb for instance, who’s been shouting that Wall Street is filled with people who don’t know what they’re doing for quite a while now) than birthdates.

  24. Gravatar of JTapp JTapp
    28. April 2009 at 02:57

    “At some point the economist will make an assertion that seems to me to be in conflict with the EMH. ”

    #1 reminded me a lot of Taleb’s Fooled by Randomness. He would answer #1 by saying “Yes, at BEST.”

  25. Gravatar of ssumner ssumner
    28. April 2009 at 04:43

    These are all good questions:

    Larry, Here is a short post on stagflation:

    Stagflation does not push up all nominal aggregates, it pushes up inflation. NGDP is unaffected by stagflation. That’s just one more reason to support NGDP targeting (when an oil shock hits, you don’t have to go through the painful process of shrinking non-oil prices.)

    Most economists overestimate the importance of AS shocks. Even our textbook example of the 1970s doesn’t really fit it all that well. In our textbooks when AS shifts left it reduces RGDP. But in the 1970s RGDP actually rose, and indeed rose at a fairly normal pace.

    It is hard to find (post-war) examples of U.S. recessions caused by supply shocks–with the possible exception of oil shocks. But even oil shocks are iffy. Recall that in 2007 and the first half of 2008 our real economy kept growing despite a massive oil price increase, AND DESPITE THE SUBPRIME FIASCO AS WELL.

    In contrast let’s look at AD shocks (NGDP shocks). In virtually every case of major NGDP shocks, RGDP goes in the same direction. Late last year is a good example. But that should not be true if my opponents are correct. Just to review, my opponents say “sure you could raise NGDP by 5%, but the real shock of banking and housing distress would mean you’d simply get a lot more inflation.” Again, there in simply no evidence in U.S. macro data to support that claim. No other banking crisis (1930-33) (2007-08) seems to have had supply shock characteristics. The fact is that when the economy is very depressed, like right now, NGDP increases always lead to more RGDP in the short run, as the SRAS is fairly flat at depressed levels (obviously not totally flat.) Here is one area where I am not a crank. Most Ivy league profs who have done a lot of empirical work and who really understand the numbers would agree with me. (Although that doesn’t prove I’m right.)

    Jon, Let’s first be clear that the previous two expansions were somewhat atypical. In both cases there were substantial real bubbles that seem unrelated to economic growth. But contrary to the Austrians, those are the only two examples I am aware of in the last 100 years. The 1960s saw an very large inflationary boom—far larger than 2003-07. But there were no serious structural imbalances within the economy. The boom was very structurally balanced. Ditto for the 1920s. Now it seems to me that the Austrians play a sort of game–they look at the investments ex post and say “see, all those wasteful projects were built–like the Empire State Building built right at the tail end of the 20s boom.” But those projects only look foolish because of bad macro policy (even by Austrian standards–remember Hayek favored steady NGDP, not 50% decreases), the projects would have been perfectly fine had the economy kept growing , as it was capable of doing and as it was expected to do. Only the dotcom and recent housing bubble look like investments that would have been foolish, EVEN HAD NGDP KEPT RISING.

    I’ve actually have commenters tell me there were other housing bubbles beside 2006, even though the U.S. house market was about as non-bubble like as one can imagine—until last year not a single year of falling housing prices at the aggregate level–it doesn’t get much more non-bubble-like that that. So bubbles and imbalances are simply a horrible theory of business cycles, they tell us nothing about most cycles.
    Regarding the first part of your question, I accept the sticky wage/price explanation of why monetary policy has real short run effects. Because sticky price markets are not easy to arbitrage, the somewhat predictable price movements are not a violation of the EMH. Don’t confuse the new classical model that says nominal frictions don’t matter, with the much more widely accepted new Keynesian model, which does have Ratex, and does have the EMH, but also has stocky wages and prices for goods not in auction-style markets.

    AdamP, Let me start with the easiest point to refute. The EMH does not say it is possible to know the correct value of stocks, the world is far from a perfect information world. As we get new information we constantly revise our estimate of what they are worth. The EMH says the market (not any individual like me) comes up with the optimal estimate, given all available info. So I don’t know what stocks are worth.

    The problem with relying on stocks as a monetary policy indicator is that they do poorly in both periods of too high inflation (1970s) and too low inflation (early 1930s). So falling stock prices do not, by themselves, tell you that money is too tight. When combined with other indicators, however, they can provide useful info about the real economy. Throughout my blog I have generally tried to link stocks and commodities together, when they move sharply and in the same direction, it is often an indicator of expected real output movements. But the best market indicator is TIPS/conventional yield spreads. They showed deflation expectations in late 2008. I am saying that someone like Lucas should look at a wide range of market indicators. If he had done so, I believe he would have agreed with me, I believe he would have also thought the markets were expecting AD to come in too low. Whether he would then agree with my policy critique, is an entirely different question. He might worry too much about policy lags.

    Soren, My primary point was that I think macroeconomics as a whole implicitly violates the EMH. That is different from the question of whether the EMH is a good one. I think it is for reasons I discussed in several other posts. I think it can help us understand monetary policy and business cycles, it can help us understand why a cut in interest rates in December 2007 was actually a sharp tightening of monetary policy (see my ratex post from 2 months back). I agree with your assertion that if markets are irrational then macroeconomists would be justified in blowing off market info. But I don’t think they are irrational. They have been far more prescient than private forecasters during this crisis. The markets were crying out for much easier money last fall. Most private forecasters were not. Who looks correct today?

    I don’t agree with either your point a or b. I think most people are not smart, just as individual brain neurons are not smart. But I think investors as a whole provide the smartest view of the economy we have, just as the brain as a whole is smarter than any individual neuron. There is no person, not even Warren Buffett, who is as smart as the overall market. Even Buffett stays invested during both ups and downs in the market, not just ups.

    JTapp, Read my previous response. There’s a word for economists who think they are smarter than the markets—fools.

  26. Gravatar of Adam P Adam P
    28. April 2009 at 08:57

    Scott, first of all I never even slightly suggested using stocks as a monetary policy indicator. I’ve no clue where you got that idea.

    I was merely pointing out that your claim that “markets” disagree with Lucas’s optimistic view doesn’t seem to apply to all markets. Stocks clearly priced quiet optimisticly.

    With respect to TIPS spreads, certainly you know that TIPS spreads give a risk-neutral expected value for inflation. As you know, the risk-neutral measure differs from the physical measure by overweighting low utility outcomes. Couldn’t it be that in the physical measure inflation was expected but there was a small probability of a massive deflation and this was such an incredibly bad outcome that it had a very large weight in the risk-neutral measure? Do you have any evidence to refute this possibility?

    Finally, I don’t want to argue interpretations of the term “worth” but the EMH, as stated by Fama, certainly could be viewed as saying that we do (both of us) know what stocks are worth because the market price fully reveals their worth.

  27. Gravatar of Adam P Adam P
    28. April 2009 at 09:09

    actually I should modify my statement that stocks are clearly priced quite optimistically since that’t not what I actually think. My point was that you could certainly, perhaps naturally, interpret stock prices as quite optimistic given the valuations and if Scott believes the prices are not very optimistic he should explain his reasoning.

  28. Gravatar of ssumner ssumner
    28. April 2009 at 09:20

    Adam, I never said that you thought stocks were a good monetary indicator, in fact I assumed you thought they were not. I thought that you were saying I viewed them as a good monetary indicator. That’s why I explained why I thought they were a lousy indicator.

    Yes, your hypothetical about risk is certainly possible. It’s also possible that the fiscal stimulus package was a mistake because in 3 months we will have an overheated boom and need need to run budget surpluses. We live in a world of uncertainty and policymakers must operate with their best estimate of what is going on. As you know, almost all macroeconomists disagree with my view of the crisis, but I don’t think they disagree because they dispute my view about sharply falling NGDP expectations last fall. I had lunch with Mankiw and Ball last fall and started by trying to explain to them why I thought NGDP expectations were falling fast. Mankiw immediately said something to the effect that almost everyone (Including the Fed) knew NGDP growth expectations had fallen way below target. Mankiw then said something to the effect of “the question is what to do about it.”

    So if we want to have a debate about how plausible my views on NGDP expectations were, I’d be thrilled. I have discussed all sorts of different markets all pointing in the same direction, I have the fact that most private economists were furiously scaling back their growth forecasts last fall, and I have the fact that actual NGDP was also falling fast. I’d say that’s enough evidence for policymakers to act on in a world of uncertainty.

    No indicator is perfect—until we get my NGDP futures plan adopted (and even that one’s not quite perfect, but close enough.)

  29. Gravatar of david glasner david glasner
    28. April 2009 at 09:22

    On EMH, I agree that it is fair to ask of any model, why, if the agents whose behavior drives the model understood the model, would their behavior not be different from what the model says it is. I think that is a powerful counterargument, as if there any shortage of counterarguments, against the Austrian theory. And parenthetically I will just add that the Austrian theory has many important lessons to teach, but if you think that all you need to know about business cycles is contained in the Austrian theory of business cycles, which seems to be a cardinal tenet of the people who populate sites like the Mises blog, you are very far removed from reality. However, to reject someone’s criticism of your theory by saying, “if your criticism is valid, then why aren’t you rich?” is just a verbal ploy not a real argument. For example, how would you make that argument effectively against George Soros, who, in his annoying and self-absorbed way, has been insisting for almost 20 years that his rejection of EMH, or what he calls market fundamentalism, has allowed him to identify and capitalize on market errors as the most successful hedge fund manager in history? Are you saying that, if he is right, he should be even richer than he is?

    I actually believe that Soros has identified a key weakness in EMH, which is that EMH assumes that asset prices reflect fundamentals. I interpret that to mean that at any instant there is an equilibrium value for the asset that would reflect the aggregate of all information available and that the current price is an unbiased estimator of that equilibrium price. The problem with that view is that the information on which the equilibrium depends includes individual expectations, so for most if not all asset markets, it is not possible to identify a set of fundamentals that is independent of expectations. So if expectations change, the equilibrium conditional on those expectations also changes, so it is really not possible a priori to distinguish between bubbles and non-bubbles, though there may be extreme cases where expectations are simply not consistent with any possible world. In markets which are closely tied to physical constraints on supply, expectations in either direction may be bounded sufficiently so that fundamentals do constrain potential equilibria fairly tightly, but for many stocks, for currencies, for the price level, how are equilibria constrained by anything? Expectations rule! Keynes’s description of stock markets in the General Theory, which I used to laugh at (stock prices of sellers of ice are higher in the summer than in the winter, ha ha!) is not as off the wall as I thought. As another parenthetical aside, I would observe that a careful reading of Hayek (especially his seminal “Economics and Knowledge” paper would lead one to a similar conclusion. So Soros should be a bit more careful in drawing policy conclusions from his insight into what he grandiosely calls his theory of reflexivity as well as a bit more modest in his assessment of his own originality.

    In particular, since this blog is all about inflation and inflation expectations. If the public is convinced that we are headed for a depression and that means that there will be a deflation, we will indeed be headed for a depression and deflation. In a fiat money world, can changing the money supply change that expectation? How would it do so? By magic? It’s really another version of the mind/body problem. We seriously do not know by what laws of nature my decision to type the words that I am now typing results in the physical effect that I type the words that I am typing. It’s just a mystery. Like where did the first living cell come from? How did matter come into existence? We have no clue. That the price level of a fiat money economy is what it is, is equally unexplained. There’s just no explanation for that price level apart from the assertion that expectations about future prices are what they are. Even the assumption that otherwise worthless fiat money can have a positive real value because it can be used to discharge tax liabilities to the state doesn’t get you very far, because the value of fiat money is dominated by the monetary demand, which is entirely a product of expectations of future prices. That is exactly what Fischer Black thought, and it’s a little scary to be coming around to his view almost 40 years after he scandalized his colleagues at the University of Chicago, and especially Milton Friedman, by making that argument.

    Having said all that, I still believe that if the monetary authorities kept repeating that they were going to inflate and would keep printing more money until they got the inflation they wanted, they would get the inflation they wanted. Is that logically demonstrable in any extant monetary model? I don’t think so.

  30. Gravatar of Adam P Adam P
    28. April 2009 at 09:30

    The point about risk-neutral vs physical measure is not quite a hypothetical. We agree the world is uncertain, we need to glean useful information from very, very noisy signals and take the right action. You want to use the market prices as sufficient statistics in effect treating prices as expected values of random variables.

    However, the most basic economic theory tells you that market prices evaluate expectations in a different probability measure than the true one. Thus prices give biased estimates of expected values. Now, I think one can make a strong argument that the risk-neutral expected value is actually the one to use in policy making. I think you can also make a strong argument the other way. I haven’t heard you make either.

  31. Gravatar of Adam P Adam P
    28. April 2009 at 09:37

    Not to step on Scott’s toes but, in response to David’s statement: “Having said all that, I still believe that if the monetary authorities kept repeating that they were going to inflate and would keep printing more money until they got the inflation they wanted, they would get the inflation they wanted. Is that logically demonstrable in any extant monetary model? I don’t think so.”

    Well actually, the canonical New-Keynsian Phillips Curve pretty much explicitly says that if you raise expected future inflation the result is higher current inflation.

    Now, let’s not argue about Calvo pricing since nobody, even the authors of the canonical model, like it so much. We all agree it’s just an analytical device. Nonetheless, given the specification of the constraints in the economy the model derivation is tight.

  32. Gravatar of david glasner david glasner
    28. April 2009 at 09:51

    Adam, I was referring to the link between Fed actions and the public’s expectations of inflation. There is a link, even in a simple-minded Monetarist model, between expected inflation and actual inflation. It is the link between policy and expectations that is without foundation. The New Keynesian view, by the way, which has become the prevailing orthodoxy among post-Monetarists as well, is simply a restatement of Fischer Black’s 1970 anti-Monetarist argument which so irritated Friedman when Black dared to air his heretical views in the U of C money workshop.

  33. Gravatar of Adam P Adam P
    28. April 2009 at 10:10

    david: “It is the link between policy and expectations that is without foundation.”

    In the NK paradigm policy is the setting of the short nominal interest rate. The logic of the model is that the policy variable directly effects current inflation, thus expectations of future policy directly affect expectations of future inflation.

  34. Gravatar of david glasner david glasner
    28. April 2009 at 10:36

    Okay, you have identified a link, but not everyone would accept that it is a very tight link. And if expectations about the future price level are pessimistic, why should the zero interest rate peg change those expectations. Nevertheless, as an empirical conjecture, I believe that the monetary authorities can positively affect expectations about inflation, just as I believe that by deciding to take an action, I can will my body to follow my mind’s instructions. I think that the connections in the two cases are equally unexplained scientifically.

  35. Gravatar of Greg Ransom Greg Ransom
    28. April 2009 at 12:01

    Scott — Orange County, CA was already in a dramatic downturn in 2007. It’s empirical nonsense to insist that it began in Oct. of 2008.

    What sort of “science” lets you assume that a math construct stipulated in your own head called “EMH” can rule out misallocations of production goods and labor across the complex time structure of production, investment and consumption? Witchcraft? Supernatural forces? Black magic? Astrological powers? An all-knowing supreme being? Superman?

    Scott wrote:

    “I believe that the depression was caused by events that took place in September and October, when the markets actually crashed. Which depression? All of them—1929, 1937, 2008, etc. And as far as I know I am the only economist who believes that all of these depressions were caused by events that occurred in those two fateful months.”

    And Scott wrote:

    “At some point the economist will make an assertion that seems to me to be in conflict with the EMH. And after that point I have trouble taking anything they say seriously.”

  36. Gravatar of Greg Ransom Greg Ransom
    28. April 2009 at 12:06

    The top Hayekian macroeconomists — Roger Garrison — also has a dual causation explanation of the current boom / bust. See his FEE lecture from early 2008 (note well, a lecture Garrison gave before Oct. of 2008).

    Nick wrote:

    “My own view is a multiplicative mix of yours and the mainstream view. Multiply financial fragility times monetary policy mistakes.”

    I’ve collected dozens and dozens of articles documenting the multiple dimensions of the current crisis — the particulars of the current episode is “overdetermined” to use the language of philosophers of causation and explanation.

  37. Gravatar of Greg Ransom Greg Ransom
    28. April 2009 at 12:22

    Scott, you’d have to explain to me how the “EMH” rules out the misallocation of production goods across the time structure of the production, investment, and consumption. The “EMH” is a terribly weak and question begging assertion when your are talking about a structural money and credit distortion across every possible dimension of relative prices across time and production and investment and consumption.

    The whole question begging issue just is what counts as “knowledge” available to actors in the market in such an environment. As Hayek points out, “knowledge” isn’t a given in the market, it’s constantly discovered — something more alike an evolving judgment and perception. And it most certainly isn’t a “given” in a macroeconomist’s math construct.

    Just guessing, but I don’t think you “get” Hayek’s economics — i.e. the real world microeconomics of heterogeneous production and investment across varying time dimensions involving changes in the shifting significance of monies and credit at various particular places in that structure.

  38. Gravatar of ba feed » When You Put It That Way, Scott…, by Bryan Caplan ba feed » When You Put It That Way, Scott…, by Bryan Caplan
    28. April 2009 at 15:05

    [...] Caplan I can’t believe how much excellent material Scott Sumner hides “below the fold.”  A prime example: Whenever I read opinion pieces by almost any macroeconomist– Keynesian, monetarist, new [...]

  39. Gravatar of Joe Calhoun Joe Calhoun
    28. April 2009 at 16:02

    I agree that the events of September/October caused this depression (if that’s what we’re calling it now), but I wonder what it is that you think changed in that time frame. Many have pointed to the Lehman failure as the trigger, but that doesn’t make sense to me. The market seemed to handle the Lehman failure and the settlement of its associated credit default swaps pretty well. What is it that changed in that time to cause the crash? I understand that you believe monetary policy was too tight going back to earlier in the year, but to get a crash, it seems to me that something had to happen suddenly. So, what caused monetary policy to suddenly become so deflationary in late September? I believe that what changed was the introduction of massive uncertainty by Paulson/Bernanke/Bush.

    As an investment advisor, I had to make real time assessments of the situation and I can tell you that in the heat of the moment, that was not an easy thing to do. I came into the Fall having booked solid gains in the first half of the year in commodities and had a fair amount of cash (roughly 15%). My stock allocation was lower than normal but not seriously underweight and I had no specific exposure to financials (except through index funds). I had no exposure to real estate and I had a healthy allocation to Treasuries. So I was feeling pretty good about things – until Paulson and Bernanke panicked. There was no way to predict that. No model would have predicted it. Anyone who claims to have predicted the crash was either lucky or they’re lying.

    When Paulson and Bernanke panicked in such public fashion, the public’s attitude about what was going on changed immediately. I know because I started getting calls from clients that I have never gotten in the twenty years I’ve been investing other people’s money. I have preached to my clients for years about long term investing, portfolio construction, non correlated assets, etc. and it was like they never heard a word of it. They just wanted out of everything. I manage accounts on a discretionary basis and I was already selling stocks to reduce exposure in late September, but by early October a lot of my clients didn’t want to just reduce risk; they wanted to eliminate it. I’ve never heard fear like that from a client.

    In real time it is hard to recognize when a shift like this happens. My view of the economy going into the Fall was that the drop in real estate values would not be sufficient to cause anything more than a garden variety recession. And I still think that if Bernanke and Paulson hadn’t panicked, that is what we would have gotten. For what its worth, it was their panic – and my clients reaction to it – that changed the equation and moved me to take a darker view of the economy and the markets.

    I hope this adds to the discussion. I think a lot of times macro economists get lost in their models and forget that there is a human element to economics. Economies are about people and how they react to the signals provided by the market. No model can capture the emotion of the players on the economic stage.

    As for the other topics, some observations:

    Democrats are not knowingly socialist, but the result is the same.

    Tyler Cowen is a freak and doesn’t need sleep. When will he crash?

  40. Gravatar of ssumner ssumner
    28. April 2009 at 17:02

    David raises an interesting question, and I’d be curious as to Greg’s response. David basically says that the Austrian model assumes the public doesn’t believe in the Austrian model. At least that’s how I interpret David’s remark. In other words, the Fed does something wrong (from an Austrian perspective), and it causes people to make decisions that are not in their own self-interest. My hunch is that Austrians would not agree with this characterization of their model. But every time an Austrian leaves a comment on my blog, that is the impression I get–that they assume the public does not understand the bad signals being sent out by the Fed. Bennett McCallum said ratex should actually be called consistent expectations, the term ‘rationality’ is very misleading and really has nothing to do with ratex. It is about expectations consistent with the model.

    You are right that I shouldn’t have said “If you’re so smart. . ” maybe I’ll add an update.

    I don’t understand your defense of Soros. What would I say to him? I’d ask what use is a theory (the anti-EMH theory) that only benefits a couple people out of 6,600,000,000 people? OK, 2 out of say 100,000 wealthy investors. (I assume the other guy is Warren Buffett, who also seems to have outsmarted the markets.) Another argument is that if the EMH was completely true, just on the basis of chance you expect a couple people out of 100,000s of wealthy investors to get lucky. But let’s say Soros is right. What use is it? If only he knows it, then it doesn’t help policymakers. If he goes out and converts the investment community to think like he does, then his strategies won’t work anymore–they will be priced into assets. So that’s what I’d say to him. Does he think he can educate government bureaucrats to his mysterious secrets, but still keep them secret from investors? I must be missing something.

    I’ve never understood why economists think the value of fiat money is such a big mystery. It’s useful in transactions. People are willing to hold X% of wealth in a convenient, small denomination nominal asset that is widely accepted in transactions. And remember there are legal restrictions against private cash. The government has a monopoly. Of course the X% depends on the expected rate of inflation. But this is hardly complete guesswork. We know that the Fed is much more likely to target inflation at 2% than 200% or negative 80% for political reasons. These indeterminacy models tend to assume money is just another nominal asset like T-bills, but it’s not. When I go to the store I don’t think about whether to take cash today, or that close substitute T-bills to the store.

    I don’t think expectations create any big mysteries at all. People try to look at the world realistically. Ex post they often make mistakes, but that’s just as likely to occur with real assets like houses, as with cash.

    Here’s another way of thinking about cash. We know a gold peg can pin down the price level. But an expectation that for political reasons the Fed will try to control inflation is sort of like a gold peg, just with more goods, and a bit more uncertainty about the future average price of those goods. But even under the gold standard there was some uncertainty. In 1929 people thought the dollar would be 1/20.67oz gold five years later. It wasn’t. There was uncertainty, but doesn’t everyone agree that a gold standard pins down the price level?

    AdamP, You tell me the optimal way of generating expectations, and I’ll say let’s use that method. So we have no dispute there. Here’s my default position (but I’d love for you to improve on it.) I would like NGDP targeting. So the Fed should create an NGDP market, and subsidize trading to create liquidity. There may be a small risk premium embedded in the NGDP price, but it only causes problems if it is time varying. Here’s one reason I don’t think the risk premium would be that big–we don’t currently have an NGDP futures market. What does that tell you? That there is not much demand for NGDP futures as insurance, as a hedge. So the risk premium is probably relatively small, and the time varying part (which is all that matters in the long run) is probably smaller still. But again, if you find a better way to ascertain the optimal NGDP forecast, I’m all ears. My main point is that we should target the forecast, however derived. We should steer the tanker toward open water, not right toward the rocks.

    Greg, Orange County is not the U.S. The subprime crash started in 2007 in certain markets. A broader real estate crash started in August 2008, and occurred FOR TOTALLY DIFFERENT REASONS, and was nationwide. The first crash was due to misallocation, as you say. The second and more severe one was caused by the sharp fall in NGDP, in other words by a very tight monetary policy. That’s what sharply falling NGDP does, it causes all sorts of asset markets to crash. Misallocation of resources doesn’t do that. It merely causes crashes in specific sectors, not the entire economy. When resources are misallocated, we should reduce output in the overbuilt sector, and increase output in other sectors, or at least, if it is hard to quickly reallocate resources, at least maintain output in non-bubble sectors. But we have a generalized fall in NGDP, exactly what Hayek said we should avoid. And you are right, I do know almost nothing about Hayek’s macro.

    But I do know a bit about the EMH, and it certainly does not say that there can be no misallocation of resources. Rational expectations does not equal perfect foresight. People make lots of mistakes, the world is very chaotic. On that we agree. But there is a different between complex root causes to the housing/banking crisis, and the simple causes for the NGDP crash. The monetary authority has the ability to control NGDP. I agree that had the subprime fiasco not occurred, the Fed would not have let NGDP drop, but even with the subprime fiasco, the Fed had the power to prevent other sectors from crashing due to falling NGDP. They took their eye off the ball. I believe Hayek warned about not letting NGDP fall.

    If we knew a bad policy was going to result in an economic crash, then the crash should have occurred at the time the bad policy was announced. If we didn’t know the bad policy was going to result in a crash, then we didn’t know enough not to do the bad policy. That’s my point.

    These were good questions, sorry if the answers seemed a bit rushed–I’m doing too many things at once right now.

  41. Gravatar of Greg Ransom Greg Ransom
    28. April 2009 at 20:42

    Scott — it doesn’t matter what “model” people believe in, where talking about the real world here, not a new classical math construct. People are competing in an environment where their competitors are making use of the money expansion and interest rates shaped in part by Fed policy. The money and the credit enters the economy at particular points. People NEVER have full understanding of the significance of all this. To begin with, it doesn’t matter to a great degree what people think is happening in terms of economic theory — the structure of production at different time dimensions of production is getting shifted “automatically” by the system wide effects of the changing value of money, the particular injection points of new money, and the immediate signals given by changes in the price of money and credit. If the housing market is in a bubble, it doesn’t matter if you know its a bubble, it’s in your self interest to ride the bubble, some way or other. And there will be no way to know all of the particular places the bubble is taking place or how it is manifested system wide. Your theory and your guesses don’t matter.

    I’ll return to this point in another post.

    Scott wrote:

    “David basically says that the Austrian model assumes the public doesn’t believe in the Austrian model. At least that’s how I interpret David’s remark. In other words, the Fed does something wrong (from an Austrian perspective), and it causes people to make decisions that are not in their own self-interest. My hunch is that Austrians would not agree with this characterization of their model. But every time an Austrian leaves a comment on my blog, that is the impression I get–that they assume the public does not understand the bad signals being sent out by the Fed.”

  42. Gravatar of Adam P Adam P
    29. April 2009 at 02:59

    Scott says “Here’s one reason I don’t think the risk premium would be that big–we don’t currently have an NGDP futures market. What does that tell you? That there is not much demand for NGDP futures as insurance, as a hedge.”

    Or it tells me that nobody has yet been willing to make a market because they can’t hedge their net futures position by trading the underlying asset. If there’s little demand for hedging NGDP then why do stocks (at the diversified index level) have such a high risk premium?

    On the other hand, suppose there is little in the way of hedging demand for NGDP futures, do you then risk having a thin market with not very informative prices?

    Finally another question. Is it desirable to have monetary policy be as volatile as the typical liquidly traded asset price?

  43. Gravatar of ssumner ssumner
    29. April 2009 at 05:35

    Greg, When people say it is in your interest to ride the bubble, I have no idea what they are talking about. The bubble term is very vague, but the only way I can make any sense of it is to assume that a bubble price is a price much higher than the expected future price. But why would it be in your interest to buy something at a price which is higher than the expected future price? (And please don’t say they plan to ride it a bit higher to the peak and then sell out–that would be an irrational belief for everyone to have, because they would have no one to sell to at the peak.)

    AdamP, The market need not take a net long or short position, it can merely facilitate trades. Markets like Intrade do fine with that approach. So the lack of a physical counterpart to NGDP futures is no problem.

    My NGDP proposal calls for the central bank to subsidize trading. But even if they didn’t, recall that they will buy or sell unlimited NGDP futures at the target price. So that makes the market very liquid, and trading would certainly be adequate if the expected future price diverged from the target price. But again, I have always favored a subsidy in the form of interest on margin accounts at a high enough rate to encourage adequate trading volume.

    Regarding the last question, the answer is no. I should also say that the last question is completely irrelevant to my proposal, as I am calling for an NGDP contract. Just as actual NGDP is much more stable that actual foreign exchange rates, so would NGDP futures be much more stable than exchange rate futures. Indeed under my plan NGDP futures would be completely stable, so asset price instability is simply not a problem.

  44. Gravatar of Jesse Jesse
    29. April 2009 at 06:21

    And please don’t say they plan to ride it a bit higher to the peak and then sell out–that would be an irrational belief for everyone to have, because they would have no one to sell to at the peak.

    It’s also irrational for everyone to think that they’re a better-than-average driver and that they have a better-than-average chance at getting rich, and yet somehow very many people manage to hold these beliefs.

    The Efficient Markets Hypothesis will hold as long as the “smart money” can spot and eliminate all the mispricings. Since in a sufficiently liquid market the “smart money” can theoretically be a single smart person, this may seem like a weak condition, but it’s really not. Lack of substitutes, risk aversion, and principal-agent problems can all contribute formidable obstacles to arbitrage. See “Inefficient Markets” by Harvard economist Andrei Schleifer.

  45. Gravatar of 123 123
    29. April 2009 at 08:04

    “Shiller lost money because he was wrong in saying stocks were overvalued in 1996.”
    1996-2009 stockmarket returns are very poor…

  46. Gravatar of Adam P Adam P
    29. April 2009 at 08:10

    Scott, your your missing the point about the volatility of monetary policy. Price volatility on your contract is not an issue since you want it fixed at the target. However, you want monetary policy to mechanicaly react to futures positions in the market, here’s how you describe the process on the worthwhile canadian initiative blog:

    “The way it works is as follows. If investors think NGDP growth will be less than 5% (like right now) they would buy NGDP futures contracts. Each time they bought one, the Fed would do a parallel OMP (probably involving T-bills) to actually change the base.”

    Thus as news comes out and flows go violently between long and short in the NGDP futures market you will have monetary base jumping around a lot from day to day. This is the volatility of monetary policy that I referred to and it would happpen. I know you’re claiming that the flows would be stable but that’s amazingly unlikely. The canonical New-Keynsian model also has this feature in theory. The Taylor principle completely stabalizes inflation. In practice of course that doesn’t happen. Are you really claiming to have the first policy in history to work just as well in practice as it does in theory?

  47. Gravatar of Greg Ransom Greg Ransom
    29. April 2009 at 08:17

    Scott — imagine you are an immigrant without an education, you are a home builder, you are a mortgage finance company — and you located in OC, California in the 2000′s, and you find the price of money and credit crashing, your production goods costs are lower, people have tons of credit to buy your product.

    In that environment you are not going to stand on the sidelines while your competitors take all your business. You are going to participate.

    Likely your almost zero understanding of”macro” tells you the good times have no reason no to go on and on. And it doesn’t matter anyway, if you have reason whatever to anticipate that things will slow down anytime soon.

    So I have no idea what Scott is talking about when Scott writes the following:

    “When people say it is in your interest to ride the bubble, I have no idea what they are talking about.”

    Is Scott denying the actual? We know OC, California in mass participated in the housing boom.

    Are we going to believe are fake / impossible / not even close to being in this universe math models — or our lying eyes?

  48. Gravatar of Greg Ransom Greg Ransom
    29. April 2009 at 08:18

    Add a “no’ in there:

    “Likely your almost zero understanding of”macro” tells you the good times have no reason no to go on and on. And it doesn’t matter anyway, if you have NO reason whatever to anticipate that things will slow down anytime soon.”

  49. Gravatar of ssumner ssumner
    29. April 2009 at 08:30

    Jesse, I was responding to Greg, who I had thought was denying that bubble like behavior was irrational. I take it that you are taking my side–that is assuming that if people acted that way they would be irrational. By the way, since we are dropping Harvard names here, you might be interested in my post from about a month back “Richard Rorty and the EMH.” I argued that the anti-EMH had no practical implications, and thus was useless. The post was highly praised by Greg Mankiw, another Harvard economist.

    123, How do the 1996-2009 returns relate to my point? Perhaps if Shiller had shorted the market for 13 years he would have made money. Remember the joke about give them a date, or a number, but never both? Are you are telling me that I should be impressed by someone saying “eventually they will be a decline in the stock market” Sorry, but history shows there have been lots of declines, unless you can tell me when, the prediction is pretty worthless. I still don’t see a shred of evidence that stocks were overpriced in 1996 (when the Dow was about 6400.) But I didn’t say stocks were definitely correctly valued in 1996, but rather he was wrong to say they were overvalued. There was no good evidence that stocks were overvalued on that date. People often use things like historical P/E ratios, which seem of little value to me.

  50. Gravatar of ssumner ssumner
    29. April 2009 at 08:44

    AdamP, I don’t care if the MB jumps around, as long as NGDP expectations are stable.
    New Keynesian models like the Taylor Rule are generally backword-looking, so there is no reason to believe they will stablize NGDP nearly as well as a forward-looking policy.

    Greg, This may be hard to believe, but I in fact don’t believe that the banks who made all those bad loans expected to lose a fortune. I don’t think they expected those loans to make their bank stocks lose 90% of their value, and destroy much of their personal wealth. If they had known this, they would not have made so many sub-prime loans, regardless of what other banks did. BTW, thousands of banks did stay away from heavy involvement in this fiasco, and are doing well today. So I don’t buy the peer pressure argument.

    Not only do I think the banks didn’t know this was going to happen, I also think the Fed didn’t know it was going to happen, and I think most academics didn’t know it was going to happen. I think these events are basically unpredictable. Of course a few people like Roubini did predict them–but I don’t see that fact as proving his (Keynesian) view of the world.

    Nor do I think Fed policy caused this to happen. There have been many periods of much more expansionary monetary policy that didn’t result in major bubbles, like the late 1960s-70s–when inflation was much higher, but housing prices never overshot nationally.

    If you think I am arguing that the good times will go on and on, you having been reading my blog. I don’t favor monetary policies that push output beyond the natural rate, or the PPF, or whatever you call it.

  51. Gravatar of Jesse Jesse
    29. April 2009 at 09:10

    The anti-EMH position has no practical implications? Here’s a practical implication: when someone argues that “market prices necessarily reflect our very best collective understanding of the fundamentals, because markets are efficient” the anti-EMH position allows for a “no, you’re wrong” response. This is a logical argument and the anti-EMH side wins.

    If it’s useless to argue that markets aren’t necessarily efficient, then why is it useful to argue the opposite? Something must be at stake.

    It’s true that the anti-EMH position in itself doesn’t tell you how to make money investing, or tell you anything about where the stock market “should” be. And the EMH is useful for most people because it highlights the fact that there’s absolutely no reason why they would be good at investing. But in a world where the global equity markets can plummet 50% in just a few months, it’s hardly irrelevant to point out that markets aren’t necessarily efficient. The fact that no one can beat the market seems a lot less impressive if the market may be off by a factor of 2.

  52. Gravatar of Adam P Adam P
    29. April 2009 at 09:10

    Scott: “New Keynesian models like the Taylor Rule are generally backward-looking, so there is no reason to believe they will stabilize NGDP nearly as well as a forward-looking policy.”

    The logic of the New-Keynesian models with Taylor rules is pretty simple, the policy makers can influence contemporaneous inflation. Thus, if the commitment to follow the interest rate rule is credible this directly stabilizes expected inflation by establishing the expected policy response. It’s the commitment to the rule that matters.

    You have the same problem, forward vs. backward looking isn’t the issue. If the commitment to hitting the target is not credible then there’s no reason to believe your idea will do any better at stabilizing NGDP. And there is nothing that makes commitment to your rule any more plausible then commitment to the Taylor rule.

  53. Gravatar of Adam P Adam P
    29. April 2009 at 09:15

    Just a quick follow up. New Keynesian models do have a fairly serious theoretical defect in that the stability of inflation comes analytically from all other paths having explosive dynamics. This is a pretty weak pillar on which to lean the most important policy design problem. It’s quite obvious that your idea suffers from the same defect.

  54. Gravatar of Jesse Jesse
    29. April 2009 at 09:40

    To elaborate a bit on the usefulness of anti-EMH: rationality is a much more “useful” hypothesis than irrationality, because irrationality can be used to explain anything, while rationality is somewhat constrained. That doesn’t mean it’s useless to point out that people are, in fact, often not rational. And the same goes for markets.

  55. Gravatar of ssumner ssumner
    29. April 2009 at 14:57

    Jesse, When you come up with something from the anti-EMH theory that will make the world a better place, or tell me where to invest my money, then let me know. Meanwhile, I have found the EMH to be extremely useful in my research in monetary economics, especially the Great Depression. This blog is full of examples.

    AdamP, I don’t agree that the credibility is the same. The Federal government has not defaulted on a single nominal debt instrument in the past 80 years (as far as I know.) And yet during that same span of time they have broken lots of verbal promises to hold down inflation, etc. I do think market participants would trust the Federal government to honor the terms of the 12 month futures contracts. Indeed I would even trust them to honor the nominal obligation in a 12-month T-bill. So I don’t regard the credibility issue as being at all similar.

    As far as new Keynesian models with explosive price level paths, I simply have no interest. Most of those models have no plausible explanation for why people hold cash. Clever economists have published 100s of such models over recent decades, and these studies have contributed nothing during this recent crisis. Unlike the Taylor rule, NGDP futures targeting doesn’t have to rely on discretionary decisions of policymakers, which as we have recently discovered are not very reliable.

  56. Gravatar of Jesse Jesse
    29. April 2009 at 17:02

    Jesse, When you come up with something from the anti-EMH theory that will make the world a better place, or tell me where to invest my money, then let me know. Meanwhile, I have found the EMH to be extremely useful in my research in monetary economics

    Scott, I had thought that the Efficient Markets Hypothesis was some sort of scientific hypothesis that should be judged by its relationship to empirical evidence from the real world. Now I realize that it is actually a sort of Cerberus, an all-in-one moral guide, investment guide, and economics research guide.

  57. Gravatar of Adam P Adam P
    29. April 2009 at 20:19

    Scott, in order to respond to questions you have to understand them. This issue has nothing to do with the government defaulting on the futures contract. Let’s try again.

    Your mechanism for stabalization works through changing the money supply. The credibility problem refers to whether the market believes the fed will stick to the regime even in really bad outcomes. Suppose we find ourselves in a situation where real GDP growth is -6% like it is now. For the 5% NGDP growth target to be maintained we’d need roughly 11% inflation, this is a really bad situation and the market might reasonably suppose that once this point is reached the fed might adjust the target. I agree that a promise of future inflation is what we need right now but the promise needs to be credible. Furthermore, since 11% inflation is certainly higher than we need perhaps the fed coming and saying 6% will be more credible exactly because it’s less painful.

    Now, I imagine you have in mind a story where the promise of inflation prevents us from ever reaching this equilibrium but that only works if the market believes the fed will carry through if we got to this point. It would be perfectly reasonable for agents to believe the fed will not carry through its inflation promise in such a bad case and that belief directly makes the bad case possible.

    Many agree, just like you, that the thing to do is promise future inflation to lower the real rate now. If the fed came out tomorrow with such a promise people may doubt that it would do everything neccessary. The same would be true in your regime, people may simply doubt the fed would keep on mechanically expanding the money suppply to hit the target, especially if we found ourselves in a situation like now, where attempting to hit the same NGDP growth target we had a year ago would require much higher inflation then may really be neccessary. People might reasonably think that sticking to your regime in that case would be pretty dumb and conclude the fed won’t do that.

  58. Gravatar of JKH JKH
    30. April 2009 at 05:28

    There would be no private sector participation in such an NGDP futures market at all, unless such participants perceived a risk that the Fed’s NGDP price peg would fail. Otherwise, the operation of such a market is a contradiction – there would be no need for private sector hedging or desire for speculation. The only reason there would be a two sided market in NGDP futures is if participants believed there was a risk that the NGDP target wasn’t credible.

    An analogy is (previous) hot money flows into China, based on the expectation that China would break the RMB peg and revalue.

    Another analogy is changes in the yield curve according to expectations for the fed funds rate.

  59. Gravatar of Adam P Adam P
    30. April 2009 at 05:45

    JKH, yes you’re quite correct. My point is slightly different (I think). My point was whether the fed would abandon the target in a bad situation. After all, if the fed today promissed 11% inflation no matter what it took would they be believed? If credibility is a problem for that promise then it can only be because generally infaltion is considered undesirable. But under Scott’s plan, if we ended up with -6% real GDP growth like we have now, the fed would exactly be promising 11% inflation. Agents might believe that the fed would adjust the target in that situation.

  60. Gravatar of Adam P Adam P
    30. April 2009 at 05:56

    I think a further problem with Scott’s plan is that targeting 5% NGDP growth is clearly inferior to targeting output and inflation seperately like a Taylor Rule would do (I’m not here arguing for an actual Taylor rule where interest rates are the instrument. We can have a different argument about whether interest rates or money supply is the better instrument).

    The standard Phillips curve has inflation determined by the output gap and expected future inflation. This specification allows for an equilibrium where real GDP growth gets stuck on zero and inflation settles on a constant 5%. Scott’s plan would make no attempt to change that equilibrium. This is roughly what happened in the seventies. Targeting a linear combination of output and inflation wouldn’t have this problem.

  61. Gravatar of ssumner ssumner
    30. April 2009 at 06:00

    Jesse, The philosopher Richard Rorty said, “That which has no practical implications, has no philosophical implications.” The problem with your argument is that there is no way to “test” the EMH, other than by it’s practical implications. Otherwise all you have is a bunch of arguments over whether the 1987 crash looks irrational or not. It doesn’t matter how it looks! What matters is what are the practical implications of each view.

    AdamP, If you are not arguing the Fed will default, then you have no credibility argument at all. Futures targeting solves the time inconsistency problem, as the money supply adjusts until the public expects 5% NGDP growth. If they thought the Fed would not carry through, they would sell more NGDP futures, until the market expectation for NGDP growth equaled the price set by the Fed. Suppose I am wrong and you are right. What would be the price of NGDP futures if the policy lacked credibility?

    Your point also addresses a totally different issue. You assume that 5% NGDP growth requires 11% inflation, which is absurd. Do you really think output would drop 6% if we had 11% inflation. What are you assuming about nominal wages, about profits, about house prices, about consumer demand?

    JKH, Your comment shows a fundamental lack of understanding of the way any auction-style market works. In any market, government-controlled or not, the market price is equal (or close to) the consensus market expectation. So yes, on average investors would expect no gain or loss in equilibrium, but that is true in any market, and yet trading occurs. Why does trading occur? Because individuals have different forecasts, and they respond differently to new information. So your observation has no validity at all.

    BTW, even if no trading occurred it wouldn’t be a problem at all if the market expected the Fed to hit its target. As soon as expectations moved even one basis point away from 5%, trading would resume and the money supply would start changing again.

  62. Gravatar of Adam P Adam P
    30. April 2009 at 06:02

    Scott, in order ot respond intelligently you need to actually read what I said. You’ve done neither.

  63. Gravatar of Adam P Adam P
    30. April 2009 at 06:12

    Also Scott, your response to JKH shows a fundamental lack of understanding of what he was saying.

  64. Gravatar of JKH JKH
    30. April 2009 at 06:15

    Adam P,

    Agreed; our examples probably all fall into the category of “failure” of a given peg in a strict sense, where failure means re-peg at least. In your example, speculators might have begun betting some time ago that the Fed would be forced to re-peg a previous 5 per cent NGDP price to a lower level by now. They would have been betting essentially that Fed OMO done to “defend” the 5 per cent level still wouldn’t have been sufficient to prevent a liquidity trap; or that targeting an 11 per cent inflation rate would be imprudent under current conditions.

    I expect though that Scott would argue (if he chose to respond to the point being made) that the counterfactual would have been different under a historic NGDP target, in terms something not as severe as a – 6 per cent real GDP at this time. Perhaps a question for Scott is what sort of long term volatility he would expect in the inflation and real GDP components of a constant 5 per cent NGDP target. Another question is would he expect long term (mild or otherwise) volatility in the 5 per cent NGDP target itself?

  65. Gravatar of JKH JKH
    30. April 2009 at 06:30

    Scott,

    Well, if you are disagreeing with me, you seem to be saying that private sector long positions or short positions in NGDP futures are for reasons other than hedging or speculating on the risk around a 5 per cent fixed price peg. Then I’d be interested in knowing what their motive for holding those positions is, assuming that risk is zero, as you apparently believe that it is.

  66. Gravatar of 123 123
    30. April 2009 at 12:14

    “I still don’t see a shred of evidence that stocks were overpriced in 1996 (when the Dow was about 6400.) ”

    Many successful market practitioners (e.g. GMO’s Jeremy Grantham) do not believe in EMH. They say that both profit margins and P/E ratios mean-revert. The case for mean reversion of profit margins is very strong, mean reversal of P/E ratios has weaker theoretical support but is also observed in practice.
    I agree with you that American policymakers should act as if EMH is true. Although I wish that Icelandic policymakers would had thought that EMH (as applying to credit markets) is seriously wrong.
    You are absolutely right in pointing a finger to the naked emperors of September/October events; alas your blog is unique in that respect. But why did Fed make those mistakes? I think they were afraid of commodity price pressures that were the big problem of last summer. EMH was guiding the Fed in September/October, but they chose the wrong market to follow (commodity market instead of non-existing NGDP futures market).

  67. Gravatar of Bill Woolsey Bill Woolsey
    30. April 2009 at 13:24

    KJH:

    Of course the reason why people will speculate on the nominal income future is because they expect that the actual value of nominal income will differ from the target. While the market expectation might be on target, the longs will expect that it will be above target and the shorts will believe that it will be below target. This assumes that the Fed is able to somehow avoid taking a position on the contract, because it might agree with those it is trading with, but just has to maintain the peg. If everyone agrees, then they will just close out their positions.

    My view is that the Fed needs to have discretion to make open market operations in bonds so that it can keep its position zero–so that the longs and the shorts are balanced in the market.

    AdamP:

    Your examples why the regime of nominal income targetting would be abandoned were unrealistic. How could we get a -6% real growth without a drop in total spending? And even if spending dropped so much, why would anyone expect that a recovery of spending back to target wouldn’t result in real output rebounding? Real output will remain 6% lower, but the price level will rise 11%. Why?

    The actual “problem” would be that if there really was some kind of massive supply side shock, say a plague that killed 20% of the working population. Real output and income would fall. If nominal income is kept growing on target, then there would be serious inflation. Yes, a nominal income target isn’t so good if large changes in the productive capacity of the economy occur.

    But, using the current situation, where nomninal income fell, to suggest that real income would stay low in the face of a recovery of nominal income is just to unrealistic to take seriously.

    The way the increased money supply would raise nominal income is by increasing spending in the economy. Now how does that result in prices rising 11% but output remaining 6% too low?

    My guess is that your model of the phillips curve with output gaps and expected inflation and your claim that it explains what happened in the seventies is just not taken seriously by Scott. I guess he (and I) don’t think it is realistic for poeple to continually raise their actual prices in the face of growing surpluses of goods. If if they do, the central bank shouldn’t do anything about it. (What exactly are they supposed to do? Cause a sharper recession so that inflation expectations go down? Or are they supposed to accomodate the inflation so that we can get some growth?)

  68. Gravatar of JKH JKH
    30. April 2009 at 14:02

    Bill Woolsey,

    “Of course the reason why people will speculate on the nominal income future is because they expect that the actual value of nominal income will differ from the target.”

    I agree. That’s what I said. If they expect actual value will differ from the target, then they expect the price peg will fail, by definition, or they are at least hedging against or speculating on that risk. There’s absolutely no reason for them to take a position otherwise.

    The fact that they hold these expectations of failure individually doesn’t mean their positions can’t cancel out, with the result that the actual value ends up on target. I didn’t say anything to the contrary.

    But the fact that they hold these kinds of expectations of failure is the only reason there can be a market in NGDP futures.

    So I believe you’re not disagreeing with me. Your last point is an interesting one, and seems quite logical as a desirable operating aim for the Fed.

  69. Gravatar of ssumner ssumner
    30. April 2009 at 17:41

    AdamP, It is possible that I didn’t understand your question, but if you don’t respond to my questions I cannot figure out what you meant. I asked a very straightforward question–what would be the price of NGDP futures if the policy had no credibility? That’s the issue I raised. Make it simple, assume the credibility problem is that the market expects NGDP to come in below target. In that case, according to your theory, what would be the price of NGDP futures? If you answer that question, I can understand better what you are getting at. I’ll make it even simpler. If you assume there is no credibility, and the market expects less than 5% NGDP growth, do you envision the price of NGDP futures contracts being 105, or some lower number like 103? That is what I am asking. Until you answer I don’t understand your question. Regarding JKH there were two ways to interpret his comment. One possibility is that he was asking what would happen if every single investor thought the Fed’s policy would be successful. In that case, no one would invest, and the policy would be expected to be successful. Another possibility is he meant the average market forecast was for 5% NGDP growth. In that case Bill’s comment is correct, a diversity of opinion will insure trades. If you look closely at my response, you will see that I dealt with both interpretations, although less so with the first. If JKH wants to offer a third interpretation that I never considered, fine. But I notice that he did not do so, so I doubt that you are right in assuming I misunderstood his question.

    JKH, I think real growth would be more stable under NGDP targeting, and inflation would also be pretty stable (especially compared to the last 4 years.)

    123, Your comments seem reasonable. By the way, I am not at all surprised that a successful investor didn’t believe in the EMH. In fact, I would be shocked if he did. It’s much more flattering to think that you are smart than lucky. And as I said earlier, although I think the EMH is good for me as an investor, and for government policy, I would not rule out a few people being smarter than the market on a few occasions. I do doubt, however, that anyone can predict the overall direction of the economy with any sort of consistency. Even Warren Buffett often losses large amounts of money. Of course Iceland’s big mistake was deposit insurance combined with aggressive international branching–a deadly combination whatever one’s ideology.

  70. Gravatar of JKH JKH
    30. April 2009 at 18:53

    Scott,

    I didn’t respond to your comment, because it was condescending crap. You don’t seem to be familiar at all with the meaning and implication of the word risk when it is used in finance discussions such as this. On that basis, I have no reason to doubt Adam was correct. I elaborated further on my interpretation to Adam and Bill Woolsey.

  71. Gravatar of Adam P Adam P
    1. May 2009 at 07:28

    Scott, your inability to understand a pretty basic point shows a fundamental lack of understanding of basic macro theory. So, I guess I’ll explain a bit to you.

    My points can be seen easily by considering the consumption Euler equation which will be satisfied by all agents. This is pretty uncontroversial since the Euler equation is nothing more than the first order condition for utility maximization over the intertemporal consumption path. But note that I’m making no claim about the direction of causality between changes in the real interest rate and changes in the consumption path.

    Now consider a situation where the full employment consumption path only satisfies the consumption Euler equation with a negative real rate. Notice that this does not necessarily mean that expected consumption growth is negative. It can come from the perceived volatility of future consumption (see top of pg.13 Cochrane’s asset pricing book http://faculty.chicagobooth.edu/john.cochrane/research/Papers/samplechapters.pdf). I’ll argue later why this is actually plausible as a description of our current situation.

    Next, suppose that we are in this situation but monetary policy has managed to maintain full employment anyway. Since the consumption path of each agent satisfies the Euler equation it follows we must have enough expected inflation to get a negative real rate. I stress that this is entirely mechanical from the three facts 1) euler equations are satisfied; 2) the full employment consumption trajectory only satisfies the euler equation at a negative real rates; 3) we are at full employment.

    From this we can immediately see potential problems with a 5% NGDP growth target. For example, suppose the required real rate for full employment is -6% and at full employment we get 2% real GDP growth. In this case supporting full employment requires at least (with zero nominal rate) an 8% growth rate of NGDP (I’ll discuss later how we get 2% real GDP growth but not 2% consumption growth).

    We can now see the credibility problem. If a 5% NGDP target doesn’t maintain full employment then the fed would be completely stupid not to change the target. Knowing all along that this might happen the market rightly doubts the fed is committed to the target.

    Furthermore, there will always be substantial uncertainty as to what growth rate of NGDP is consistent with full employment. After all, in a world where the fed adjusts the money supply to meet a nominal interest rate target the hard part is knowing what level of nominal rate to set, hitting the target hasn’t proved that hard. Your proposal does absolutely nothing to resolve the hard part of monetary policy.

    Now, is this situation at all a plausable description of our current situation? Well, we had been on a large consumption binge financed on credit. Basically the Chinese were sending everything they could manage to produce over to the US in return for credit. Presumably at some point they would start spending that credit buying stuff from us and at that point some part of our output would no longer go to domestic consumption, thus consumption in the US wouuld be a smaller fraction of our output. However, this is not to say people were irrational in their consumption. They probably rationally believed that future output would grow enough to sell some output to China and still maintain some consumption growth here.

    But then suppose something happens like a large spike in the prices of food, oil and just about every commodity. Sound familliar? Remember all those $250/barrel oil price forecasts. Now people mark down their output forecast and, since they’ll still owe a fraction of that output to the Chinese, they correspondingly mark down their consumption growth expectations. At the same time they probably also feel more uncertain about future consumption growth (perceive it as more volatile). We now have a situation where the current conumption path can only be supported with a negative real rate and if 5% inflation isn’t enough we get a recession.

    To conclude, these last two paragraphs also give a situation where full employment might yield positive GDP growth but negative consumption growth because the amount of output that goes to China to pay off our debt is more than the growth rate of output. People are employed but their own consumption still declines.

  72. Gravatar of Adam P Adam P
    1. May 2009 at 08:37

    Actually, while I’m on the topic I guess I should address a similar but slightly different way the fed’s credibility matters because this also explains why, contrary to Scott’s claim, the 5% NGDP target can still be missed.

    Again suppose that we have an equilibrium real interest rate that is negative. Through the futures market mechanism the money supply starts to increase. Does this have an immediate effect on aggregate demand? Not if agents believe the fed might adjust or miss the target.

    In this situation the required real return to holding a risk-free asset is negative, say -3%. If cash is yielding say zero (in real terms) then cash is the best investment you’ve ever heard of. A 3% premium with virtually no risk, if people expect deflation then you even get a positive real return to cash.

    Now, in Scott’s story this is supposed to be prevented by people never believing that the target might be missed so they immediately mark up their inflation expectations as real GDP starts to fall which lowers the real return to money and hopefully gets them spending. Will it work?

    Not necessarily. First of all if the equilibrium real rate is -6% then the fed just can’t hit the target, people still hold money as a great investment (for the 1% real premium) and we end up in a deflation that just makes things worse (the deflationary spiral).

    Furthermore, once we realize that the target can be missed it becomes self-fulfilling. If the target is feared to be missed you don’t get the increase in aggregate demand (not right away) and so you can have a sharp drop in real GDP like right now, our current real GDP growth is -6%. contrary to what Scott and Bill said in responding to my earlier comment without reading it (I never assumed 11% inflation), with -6% real GDP growth it would require 11% inflation to hit the target. But actually, we only need to promise 6% to get full employment back, and since extra inflation (especially this much) would be painful people reasonbly expect the fed to forget about reaching 5% NGDP growth for the year and instead adjust the target.

    But, since agents rationally expect the fed to adjust the target they don’t immediately mark up their inflation expectations in the face of the initial fall in real GDP, thus extra money doesn’t increase aggregate demand, real GDP continues to fall and we reach the point were the fed really should adjust the target.

    Thus, a credible commitment from the fed is just as much a problem here as it would be inder an infaltion targeting regime where the fed just announces a 6% inflation target.

  73. Gravatar of ssumner ssumner
    1. May 2009 at 17:51

    Adam P, Anything is theoretically possible. But even if your hypothetical was plausible, it would not have any implications for policy credibility. Obviously the central bank will choose the policy target they prefer, if they prefer 8% NGDP growth, then that’s the target they will choose. In any case, you still haven’t answered my question of what would the price of NGDP futures be if the policy had not credibility? It is a simple question, would they be priced at 105 or not?

    Obviously if there was a severe supply shock it is theoretically possible that real GDP could fall 6% and prices could rise 11%. Will something like that happen in the foreseeable future? I doubt it. Does it have any application to the current situation? Obviously not as oil prices were falling fast long before NGDP growth expectations plunged in late 2008. In any case, if there were a severe supply shock I would still favor NGDP targeting, if nominal wages were well-behaved, as I believe it is wage instability, not price level instability, that generates macro instability.

    BTW, I just reread your comment and you did suggest that a 5% NGDP target could produce 11% inflation–your numbers not mine.

    In any case I cannot address your point unless you answer my question as to what the price of NGDP contracts would be if the policy were not credible.

    And one other thing, since you mentioned Cochrane’s name you might be interested in knowing that he thinks my NGDP futures targeting idea is feasible.

    I really don’t know what to say to JKH, Here is the exact comment I was responding to:

    “There would be no private sector participation in such an NGDP futures market at all, unless such participants perceived a risk that the Fed’s NGDP price peg would fail. Otherwise, the operation of such a market is a contradiction – there would be no need for private sector hedging or desire for speculation. The only reason there would be a two sided market in NGDP futures is if participants believed there was a risk that the NGDP target wasn’t credible.

    An analogy is (previous) hot money flows into China, based on the expectation that China would break the RMB peg and revalue.

    Another analogy is changes in the yield curve according to expectations for the fed funds rate.”

    OK JKH, Now you say I totally misrepresented your view. All you were saying is that individuals who expected NGDP would come in right on target would not bother speculating. You weren’t claiming anything else, you weren’t claiming that that this was some sort of “problem” for NGDP targeting. All I can say it you have a very strange way of expressing yourself if what you now claim—that all you were saying is that people who think Google stock is worth $247 dollars don’t bother buying or selling Google stock when the price is exactly $247. So that was the only point you were making? You weren’t hinting that this fact is somehow a problem for my proposal?

    Fine. I’ll let others decide what they think.

    And by the way, in economics when people say a macro policy is not “credible” (your words) it doesn’t mean one individual thinks NGDP will be above or below target, it means the market forecast is for the policy to miss its target. So your comment makes no sense unless you are referring to a situation where every single person expects exactly 5% NGDP growth. But in that case, there would be no “failure,” so your comment would be wrong in any case. Unless you can come up with a better explanation for that comment, I will continue to argue that you don’t know how auction-style markets work. Feel free to explain your comment, it might be more productive that calling my responses “crap.”

  74. Gravatar of Adam P Adam P
    1. May 2009 at 22:18

    Still missing the point, the reason I ignore your question about the price of the future is it’s irrelevant. Why is it irrelevant? Because, my example is a case where the price of the future is 105, just as the fed targets, but we don’t get 5% NGDP growth in the actual economy.

    Basically my examples are examples of a situation where one of your explicitly stated assumptions fails. On the Canadian blog, when describing your proposal, you said (and this is a cut and paste quote to be sure it’s what you said):

    ” For this to work there must be some monetary base capable of producing 5% expected NGDP growth.”

    In these examples, that is not true. In cases where the equilibrium real interest rate is negative there is no level of the base today that generates ANY growth in NGDP. That’s why I modified it on that blog to say that there must be A PATH for the monetary base that delivers the growth. Even this was sloppy on my part, I should have said “expected path”.

    Now, the path that generates 5% NGDP growth might be really undesirable and thus agents rationally doubt it will truly be followed.

    And again I would suggest you actually read what a comment says before responding. In these examples it’s not a supply shock of any kind that generates -6% real GDP growth, it’s a demand shock.

    You are consistently blaming the fed for letting expected NGDP growth fall but in these examples there is nothing the fed can do in terms of today’s policy to stop expected NGDP growth from falling. In these examples the ONLY way to stop expected NGDP growth falling is to make credible commitments to future actions and thus your idea has the same credibility problems as vrtually every other policy proposal.

    Thus, I’m not saying your idea is not feasible. I’m saying it’s not helpful.

  75. Gravatar of bob bob
    2. May 2009 at 05:00

    I think JKH had a point there.

    Given that the price of a NGDP futures contract dated one year out would be static at $105 (since the Fed buys and sells on demand an unlimited amount of futures at that price) I can’t really see what the incentive would be to participate in such a market. Why not just hold $105 in cash? What’s the difference?

    If it is subsidized through interest paid on these NGDP futures accounts, where is the incentive to buy or sell if the price stays the same? You earn the same amount of interest on your account regardless of whether you are buying or selling NGDP futures, so why would participants put any thought or effort into buying instruments that never change price.

    I don’t see how a completely fixed market can provide you with any of the useful information that a normal futures market would provide.

  76. Gravatar of Adam P Adam P
    2. May 2009 at 05:41

    bob, as I understand Scott’s proposal this is not such a problem. The futures price stays at 105 because the fed only offers contracts with that price. If you take one side, say that NGDP growth will be too low (let’s call that going short), then at the end of the year you receive a payoff equal to::

    105 – realized NGDP growth.

    So, you might make or lose money depending on realized NGDP growth. The fed then mechanically changes the money supply, if the flows are net short the fed increases the money supply and keeps on increasing until the flows are net zero. If the flows are net long then the fed contracts the money supply and keeps doing it until the flows balance.

    That, if I’ve understood correctly is Scott’s proposal.

  77. Gravatar of Adam P Adam P
    2. May 2009 at 06:09

    Correction, the payoff to a short position is:

    105 – (100 + percent NGDP growth realized)

    and you multiply that by the notianl traded. I asume it was clear what I meant.

    Also bob, I believe JKH was saying something different. Suppose I think that the current stance of monetary policy will only produce 2% NGDP growth, does that mean I trade? Well no, not if I think the market will work as it’s supposed to and the money supply will soon adjust in the right way to hit the target. I can perfectly well think the current money supply will miss the target but still not bother trading because I don’t believe I’ll make any money.

  78. Gravatar of Adam P Adam P
    2. May 2009 at 06:21

    And also, I should mention that Bill Woolsey apparently also refuses to read comments before responding. Bill says:

    “How could we get a -6% real growth without a drop in total spending? And even if spending dropped so much, why would anyone expect that a recovery of spending back to target wouldn’t result in real output rebounding? Real output will remain 6% lower, but the price level will rise 11%.”

    I never said that we’d get -6% real GDP growth and 11% inflation. I said IF we got -6% real GDP growth THEN we’d need 11% inflation to hit the target.

    Now, apparently both Bill and Scott are quite sure that under Scott’s proposed policy we’d never get the -6% real GDP growth in the first place. My point is that if the fed is not willing to commit to giving us 11% inflation IF that ever happened then this lack of credibility actually makes the -6% real growth a possible outcome.

  79. Gravatar of bob bob
    2. May 2009 at 07:15

    ah thanks, I didn’t realize that the Fed would be making up the difference vs. actual NGDP at expiration. That makes sense.

    I agree with the point you and JKH are making that in order for speculation or hedging in that market to be worthwhile, the Fed would have to be pretty bad at meeting its target. If the Fed is successful, then there would be little reason to trade, correct? And any time you trade, the Fed would be turning around and working to make sure your trade is unprofitable.

    I think I see what you mean in saying that it comes down to a credibility issue if the Fed has to resort to extreme measures in order to meet a 1 year NGDP target, but that would assume that the credit channels are in working order. What about situations like now, when the banks are unwilling to lend and expansion of the money supply fails to increase prices? ie in Nick’s recent post:
    http://worthwhile.typepad.com/worthwhile_canadian_initi/2009/04/bad-banks-and-the-effectiveness-of-fiscal-and-monetary-policies-on-ad.html

    I don’t see how the futures targeting regime would work better in the current situation if it is using the same instruments as regular monetary policy. My guess is that Scott thinks that a penalty on reserves would solve this.

  80. Gravatar of JKH JKH
    2. May 2009 at 08:05

    bob, Adam P,

    See if I can get a quick and dirty comment in here, without being castigated by the host for my fundamental lack of understanding.

    I think what the three of us are saying probably converges to a similar view of things.

    Adam, there will be a cash settlement at the maturity of the futures contract equal to the differential between the contract specified GDP (say 5 per cent) and the actual realized GDP. This is the rough mathematics. So if the target is 5 and the realized is 3, there must be a cash settlement for 2. I would define it longs get paid 2 and shorts pay 2, but that’s a matter of contract convention. I think you’re looking at it the other way around in terms of directional convention. But I think we are more or less looking at the substantive result in the same way?

    Bob, my point was that nobody would take a position if everybody expected zero risk to actual GDP coming in at the target of 5; i.e. if everybody expected 5 and that there was no risk of something other than 5. If there is risk, then people starting hedging and speculating with futures positions. And if there is risk to the target of 5, by definition somebody must think there is a risk that the government will FAIL to produce an actual GDP of 5 against the time horizon of that particular futures contract, regardless of the amount of OMO it does in an attempt to steer actual GDP toward 5. And if the government fails to generate that convergence at the maturity of the futures contract, the contract when realized GDP is 3 for example will pay off to the longs the way I’ve defined the longs as per my comment to Adam above. I think all of that is roughly in line with your second comment above.

    P.S. be alert in using the word “credibility” and it’s derivatives in these parts; it’s like holding a live hand grenade.

  81. Gravatar of JKH JKH
    2. May 2009 at 08:08

    Professor Sumner,

    “Unless you can come up with a better explanation for that comment”

    I made “that comment” at April 30 5:19 as a reference to Adam P’s immediately preceding comment, where he said “The credibility problem refers to whether the market believes the fed will stick to the regime even in really bad outcomes.” Adam P understood my reference immediately, because he responded with agreement as qualified 17 minutes later, to which I responded with further explanation shortly after that. You interrupted the sequence in the middle, apparently oblivious to two commenters exchanging views. Adam P immediately responded then that he understood my point, and that you did not. The two of us communicated effectively. And that’s all I really care about; not whether you understood it, because it wasn’t directed to you.

    The sequence in question, from April 30 05:19 to April 30 06:15, includes the exchange between Adam P and me. The point I was making is inherent in that exchange, and understood by the person with whom I was having the exchange. Professor, I’ll reinforce Adam P’s repeated observation. Try reading the comments. It might restrain some of your defensiveness.

    As far as my use of the word “credible” is concerned, I was merely following up on Adam’s use of the same word in his preceding comment. We exchanged views on mildly varying interpretations of the subject viewed broadly in context, rather than nitpicking on the meaning of a word. As far as your insistent hijacking of the word “credible” for your own purpose in your world, it does remind me of Moldbug’s comment on your Austrian post: “You actually don’t seem to know anything about economics.” In some ways, it’s an implicit subset of that comment. But apart from that, your restriction on the use of the word “credible” is nonsensical and silly. Adam P makes sense of this and other matters in the general thrust of his thoughtful comments and questions regarding NGDP futures, here and on the WCI blog, and is contributing a public service in that regard, so there’s no need for me to duplicate or explain it further here. Even if I were to disagree with him on some point(s), I’m confident he works out his position with intellectual cogency, honesty and thoroughness.

    “Your comment shows a fundamental lack of understanding …”

    That’s the second time you’ve directed those words my way. If I were to think that you were ignorant of the functioning of the Federal Reserve and broader banking system, I wouldn’t come out and say it in a quite so blunt and insulting way, particularly using the word “fundamental”. In fact, I think there’s an enormous amount you don’t understand or ignore about central bank operations and the financial system that could otherwise be useful to your work. The style reveals the substance. But I think I know why you respond the way you do.

    As far as I can recall, I’ve never before used the word “crap” in responding to a blogger’s response before, or anything near it. You are the first.

  82. Gravatar of Adam P Adam P
    2. May 2009 at 08:47

    bob, JKH,

    Yes, I think there’s general agreement here. Bob and I give examples where increasing the money supply simply doesn’t cause an increase in NGDP growth and so the fed just can’t hit the target. I give Krugman’s example, bob links to Nick’s example. JKH simply says their must be SOME example because otherwise nobody trades.

    Now that I think about it though, JKH’s point is really the one that kills Scott’s story (sorry I was too busy trying to get him to understand a simple example). Here’s why JKH’s point is so fatal:

    On one hand, it is essential for Scott’s idea that people generally expect (at least with high probability) that the target will be hit. It’s these expectations of continued positive NGDP growth that prevent aggregate demand from falling. This is what prevents us, in Scott’s story, from ever having the real GDP growth of -6% that is currently going on. But if people, on average, believe enough in the target being hit to not change their consumption then they must believe enough in the target not to trade the future. So where does Scott get his liquid market?

    On the other hand, suppose people think there is enough of a chance that by year’s end the target will be missed that they bother to trade on the futures market. Well, if they bother to trade the future they probably put enough probability on the target being missed to also change their consumption pattern. Thus, the market is liquid but the plan doesn’t stabalize real aggregate demand and so what’s the point?

    Scott, you can’t have it both ways.

  83. Gravatar of ssumner ssumner
    2. May 2009 at 11:15

    Bob, Adam and JKH, Yes, I agree that if everyone expects on target NGDP growth, no one would buy and sell NGDP contracts, but that has nothing to do with policy credibility. That is true of any market. If everyone thinks $247 is the fair price of Google stock, it will move there and transactions will dry up. But I have two comments:

    1. How likely is it that everyone will have exactly the same expectations regarding NGDP growth?

    2. Even if it were true, that would mean everyone thought the monetary base was at a level where on target NGDP growth would result. In that case you don’t need any public participation, you would be expected to get on target growth with the existing monetary base.

    Now of course it is much more likely that people will at some point come to believe the monetary base is too low or two high to produce on target NGDP growth. When that perception develops then people will start buying and selling NGDP futures–as required by the policy.

    Of course in the real world people have heterogeneous views of velocity. For the purposes of this example define velocity as next year’s NGDP over this years monetary base. Futures market participants will have different estimates of that ratio, and like in any other market it is differences of opinion that leads to trades.

    The reason I thought you guys don’t understand auction-style markets, is that the arguments you used would seem to apply equally well to any market. The fact that investors as a whole believe X is the fair price, doesn’t mean no trading occurs when the price is X. So even if Fed policy is 100% credible in the sense that the market expects the Fed to hit its target, investors would still actively trade the contract. If I am wrong, and I misunderstood the point you are making, I will apologize for saying you don’t understand those kinds of markets.

    Adam, You said “-6% like it is now” that surely implied you thought that more monetary expansion now would leave real growth unchanged at -6% and instead lead to high inflation. The qualifier “like it is now” is what I responded to. I think it was a very natural assumption. otherwise, what is the point of your example, if it doesn’t even apply to the point you were making?

    JKH, You are right that there is a lot I don’t know about central banking. There is nothing wrong with not understanding something. Very few people understand much of any economics, and even I don’t understand significant parts of the field. But perhaps I should stop using the term “fundamental misunderstanding.”

    Here is a quote that shows a lack of complete understanding, in my view:

    “And if there is risk to the target of 5, by definition somebody must think there is a risk that the government will FAIL to produce an actual GDP of 5 against the time horizon of that particular futures contract, regardless of the amount of OMO it does in an attempt to steer actual GDP toward 5.”

    You imply that if person X thinks the Fed will miss its target, that they think this is true regardless of the amount of OMO they do. Not so. It merely implies they think the Fed will not actually do the right amount of OMOs, which is very different. In other words they disagree with the market’s estimate of the future NGDP/current MB ratio discussed above. So I still don’t agree with you. You are still missing something fund . . ., something not that important, but still relevant to my proposal.

    BTW, You have all sorts of praise for Adam’s thoughtfulness. Well he is smart, but I don’t see how him calling my arguments “incredibly weak” is any different from me using “fundamental lack of understanding.” But perhaps that’s just me being defensive.

  84. Gravatar of Adam P Adam P
    2. May 2009 at 11:28

    Scott says:”Adam, You said “-6% like it is now” that surely implied you thought that more monetary expansion now would leave real growth unchanged at -6% and instead lead to high inflation.”

    I mean really, what’s so hard here. Do I think monetary expansion now would leave real growth unchanged? Yes.

    Do I think monetary expansion now would lead to higher inflation? NO.

    If the equilibrium real rate is negative and people don’t expect inflation then money is held because it’s the best investment you ever heard of. Your willing to hold a risk-free asset at negative real return, you have money as being the closest thing available to riskless (of course in real terms it’s not quite riskless) and it gives you a return of right around zero.

    Money is offering a higher real return than is required to induce people to hold it. The result? They hold it, they don’t spend it, they hold it.

    How is this unclear and how does your futures targeting idea solve this? It comes directly from the first order conditions for utility maximization. Can monetary expansion make people stop maximizing their utility?

  85. Gravatar of Adam P Adam P
    2. May 2009 at 11:31

    Let me repeat what I said before:

    Basically my examples are examples of a situation where one of your explicitly stated assumptions fails. On the Canadian blog, when describing your proposal, you said (and this is a cut and paste quote to be sure it’s what you said):

    ” For this to work there must be some monetary base capable of producing 5% expected NGDP growth.”

    In these examples, this condition fails. In cases where the equilibrium real interest rate is negative there is no level of the base today that generates ANY growth in NGDP.

  86. Gravatar of Adam P Adam P
    2. May 2009 at 11:34

    Sorry I should also add to the last comment:

    There is an expected PATH of monetary expansion that can generate NGDP growth.

    However, in these examples the ONLY way to stop expected NGDP growth falling is to make credible commitments to future actions (which may be painful) and thus your idea has the same credibility problems as vrtually every other policy proposal.

  87. Gravatar of Adam P Adam P
    2. May 2009 at 12:10

    Scott as for the first 5 paragraphs or your most recent response (right up to the point where you address me directly) your still not understanding the point. I’ll repeat:

    Suppose I think that the current stance of monetary policy will only produce 2% NGDP growth, does that mean I trade? Well no, not if I think the market will work as it’s supposed to and the money supply will soon adjust in the right way to hit the target. I can perfectly well think the CURRENT money supply will miss the target but still not bother trading because I expect the money supply to change in just the right way to hit the target.

    The point is there may be a very wide range of different views about what level of monetary base will deliver 5% NGDP growth. Does that mean anybody trades? Well, what if everyone disagrees on what the right level of monetary base is but everyone agrees that the market will ultimately get it right? That is, everyone agrees that the market will eventually adjust the base to hit the target. Then everyone expects to make no money trading the future. So do they trade?

    Now your Google example is just plain beside the point and also, by the way, incorrect. You say: “If everyone thinks $247 is the fair price of Google stock, it will move there and transactions will dry up.” Well in the standard frictionless model everyone has the same information (no private signals) and so everyone agrees on the price of Google stock but there is still trading.

    Why do people trade if they agree the stock is correctly priced? Because they are buying a real asset that will (eventually, perhaps not yet) provide an income stream (either dividends or buybacks). People buy more or less of the stock depending on the probability distribution of that future income (its risk characteristics). You know, I buy more if it has small covariance with my expected consumption, that sort of thing. But, they trade despite all agreeing that $247 is the correct price.

    Now, your NGDP futures are not real assets like Google. They are contingent claims with a fixed expiry, if people don’t expect to make a profit and don’t use them as a hedge (you’ve argued the risk premium would be small) then there is no reason to trade them.

  88. Gravatar of JKH JKH
    2. May 2009 at 18:48

    Professor Sumner,

    “Yes, I agree that if everyone expects on target NGDP growth, no one would buy and sell NGDP contracts, but that has nothing to do with policy credibility. That is true of any market. If everyone thinks $247 is the fair price of Google stock, it will move there and transactions will dry up.”

    It is obvious that risk capitalized in individual trading positions in NGDP futures depends logically on the individual’s perception of policy credibility. If somebody believes the policy will fail, that person is in a position to conclude there is an exploitable risk and take a position in the futures accordingly. I have described the nature of this risk relationship in simple terms and ad nauseam in previous comments. If you don’t understand this, you don’t understand risk, which is what I’ve previously suggested is probably the case.

    You keep bringing up this false and irrelevant comparison with Google. If somebody believes that Google will fail to meet next quarter’s earnings, that person may conclude there is an exploitable risk and take a position in the stock. The NGDP policy we are discussing here and the perceived risk around it have nothing to do with Google stock. That is, unless you want to pursue some off course argument to that effect, as another distraction from the relevant point.

    That said, a given individual’s perception that there is a risk of NGDP policy failure is good enough reason to take a position in NGDP futures. The way in which Google stock is priced has nothing whatsoever to do with this decision. Futures and stock are both financial instruments. Does that also negate an argument about the role of perceived policy risk in formulating individual strategies in NGDP futures? Does the fact that the sun rises over both types of traders lend even more weight to your case?

    You’ve said nothing at all that contradicts the accuracy of my observation that individual NGDP futures positions logically depend on how those individuals perceive NGDP policy risk. I have no idea why you go on about Google and other things in an attempt to refute or distract from this rudimentary linkage.

    “How likely is it that everyone will have exactly the same expectations … you would be expected to get on target growth with the existing monetary base?”

    This is obvious. I have no idea why you fall back on these ideas, because they reinforce the correctness of my statement that individual participation in the futures market must be linked logically to individual perception of policy risk. I’ve never said I expected zero perceived policy risk by all individuals in such a market. Why would I? And I’ve never denied that the hypothetical presence of such zero perceived policy risk means no trading is required. Why would I? In fact, quite the contrary is the case. This is another distraction and false inference.

    “The reason I thought you guys don’t understand auction-style markets, is that the arguments you used would seem to apply equally well to any market. The fact that investors as a whole believe X is the fair price, doesn’t mean no trading occurs when the price is X.”

    Your argument again is based on false inference and accusation. I’ve never said anything one way or the other about the logical implications of how “investors as a whole” come to their aggregate position in the market. You’re the one that keeps obsessing on this point. Exactly the opposite is the case. I’ve consistently made an argument about risk as perceived by individual investors, an argument deliberately constructed to be independent of this aggregate perspective that you keep bringing up.

    “You imply that if person X thinks the Fed will miss its target that they think this is true regardless of the amount of OMO they do. Not so. It merely implies they think the Fed will not actually do the right amount of OMOs, which is very different.”

    This is logically absurd and incredibly picayune as well. It doesn’t imply this at all. First, not all thinking people automatically formulate their strategy by adopting your monetary base determinant for nominal GDP expectations. This is only mandatory in your restricted thought world, in which you believe every human being capable of futures trading calibrates their GDP expectations exclusively according to an OMO determinant. Second, others that do believe as you do may have an expectation for the OMO implementation path and the risk around that expectation, as reflected in my phrase “the amount of OMO it does”. But you cherry pick the word “regardless” as something that falls outside of the individual’s risk assessment, rather than applying it in reference to the size of the expectation and the risk around it, as I intended. Those who understand risk would intuitively appreciate my intended meaning. This is a desperate reach on your part.

    Please keep Moldbug’s perspective on truth in mind, professor. This subject of economics and markets isn’t yours alone. It’s ours, meaning all of us who are interested in the subject matter, and who like to explore it with somewhat open minds and civility in the exchange of ideas and experiences relating to it. This is only your blog, not your subject.

  89. Gravatar of 123 123
    3. May 2009 at 02:51

    Adam P,

    Just like there is a trade in inflation swaps now, there would be a trade in NGDP futures (swaps). Just like inflation targeting does not achieve 100% exact results, NGDP targets sometimes would also be missed at least by some amount.

    Although the best way to create NGDP futures market is to issue NGDP linked treasuries.

  90. Gravatar of JKH JKH
    3. May 2009 at 03:46

    Adam P,

    Thanks for your 12:10 above.

    Like I said before, it’s a public service for somebody with a clear mind and writing style to translate this particular subject patiently and painstakingly and correctly into understandable syntax.

  91. Gravatar of Adam P Adam P
    3. May 2009 at 04:08

    123, yes I never once disagreed with that statement. The argument is over Scott’s attempts to have it two mutually exclusive ways at once.

    Consider the logic of Scott’s claim that the fed caused the current mess by allowing expected NGDP growth to fall. The argument is that what caused the fall in aggregate demand is a fall in expected NGDP growth which causes people to attempt to save more now. On this I think Scott is quite correct. Now, does his futures targeting proposal help anything?

    Well, if agents believe with high enough probability that the growth target will be at least nearly hit then it works. That explains why when talking about stabilizing aggregate demand Scott stresses the efficiency of markets and the idea that aggregate expectations must be for the target to be very nearly hit.

    However, if agents in aggregate put such a high probability on the target being nearly hit then there will be little trading in the futures market and thus it will be uninformative. This is not actually where I think the main problem is though. I was arguing this in the context of Scott’s other claim, that the risk premium would be small.

    The real problem I see is about risk (and I think really this is also what JKH was getting at). Suppose my subjective probability distribution on end of year realized NGDP is such that in expectation the target is hit (that is the mean of the distribution is 5%). However, suppose my subjective distribution has a large amount of variance around that mean, (and perhaps a large left tail). Furthermore, suppose I’m risk averse. Then, even though in expectation I expect the target to be hit, I still reduce my expenditures. If everyone perceives a large risk like I do then the regime fails to prevent the large fall in aggregate demand and output.

    Now, one possibility is that if I saw enough risk of a fall in NGDP growth to cut back on my own expenditures I would probably just use these futures to hedge myself. Then, if NGDP falls I get paid off and thus have no need to save more now. So, if everyone does this as well does that mean Scott’s idea becomes a good one? NO, now we have a large risk premium in the futures market. Moreover, the risk premium itself will likely have a lot of time variation.

    So there, in some more detail are the two mutually exclusive things that Scott needs to both be true.

    On one hand Scott wants the aggregate market subjective probability distribution for realized NGDP to have so much of its mass near enough to the target that:
    1) Few people feel the need to reduce expenditures and save more;
    2) Few people feel the need to use the futures as a hedge;

    At the same time, even though few people expect to make (or lose) much money trading the future there is still a liquid market? Let me say it again:

    If there is enough uncertainty around realized NGDP growth to generate a liquid market then there is enough uncertainty to create a hedging demand.

    This last paragraph addresses your point 123, in the inflation swaps market there is enough uncertainty around inflation targets being hit that you have trading BOTH for profit and hedging. Scott’s idea only works if the futures are traded for profit BUT NOT for hedging.

  92. Gravatar of ssumner ssumner
    3. May 2009 at 05:32

    Adam, Here is what you said:

    “On one hand, it is essential for Scott’s idea that people generally expect (at least with high probability) that the target will be hit. It’s these expectations of continued positive NGDP growth that prevent aggregate demand from falling. This is what prevents us, in Scott’s story, from ever having the real GDP growth of -6% that is currently going on. But if people, on average, believe enough in the target being hit to not change their consumption then they must believe enough in the target not to trade the future. So where does Scott get his liquid market?”

    This statement shows a TINY lack of understanding about how markets work. In any auction-style market the average trader will believe the equilibrium price is correct, but each individual trader will have different views. It’s those differences of opinion that account for trading. If there are no differences of opinion then trading doesn’t occur, but that is appropriate because it means the current money supply is expected to produce on target growth.

    The flaw is in your language. You use “the people” in two ways; collectively, where they expect on target growth, and individually where they don’t. Imagine a bell curve of velocity expectations, with the midpoint consistent with on target growth.

    I didn’t address your earlier argument about monetary policy ineffectiveness with negative equilibrium interest rates. That only makes interest rate tools ineffective, it certainly does not make monetary base, or exchange rate, or NGDP targeting rules ineffective. There is no serious economist in the world that thinks a yen dollar rate of 1000 would have been ineffective in moving Japan out of the liquidity trap a few years ago. The ineffectiveness is about ordinary OMOs involving T-bills to target interest rates. (And even those are only ineffective if expected to be temporary, as Krugman has observed.)

    If as you say, you are assuming a price of 105 when there is a lack of credibility, then you are assuming an implausibly large risk premium. So that is my other answer, beyond what I mentioned above.

    Here’s my next question to you. People used to make the same discredited liquidity trap argument that you are making about people holding money rather than spending it in the late 1990s regarding Japan. Then a bunch of famous economists (including Bernanke) showed the argument was bogus, they showed a foolproof way out of the liquidity trap, just increase the supply of yen enough to depreciate the currency sharply enough to cause inflation. Do you not buy their argument? It seems like you don’t as your comment seems to deny that any increase in the base, not matter how large, would increase NGDP.

    Regarding Google stock, you said:

    “Now your Google example is just plain beside the point and also, by the way, incorrect. You say: “If everyone thinks $247 is the fair price of Google stock, it will move there and transactions will dry up.” Well in the standard frictionless model everyone has the same information (no private signals) and so everyone agrees on the price of Google stock but there is still trading.”

    It is important to distinguish between models that have no relevance for reality, and reality itself. People don’t have the same view of what Google is worth. So the frictionless perfect information model is beside the point. In any case, I don’t care if everyone has the same view that monetary policy is right on course. I presume they wouldn’t have that view unless it was. And if the money supply was below what they thought it would be, then the next day they would notice and trade. I foresee trading occurring every day on a new 12-month contract. So they would quickly learn if the actual money supply didn’t move to where they wanted it to be.

    By the way, my NGDP contract is just as much a real asset as Google stock. $247 is the nominal price of a real company. NGDP futures price is the nominal price of real actual future output.

    JKH, I would simply repeat what I said to Adam, particularly the first part of my reply. It applies equally well to your reply.

    123, A Treasury-linked contract would be better than nothing, but far inferior to a NGDP futures contract. Just as the TIPS indicator is far inferior to a CPI futures, as TIPS/conventional spread also includes liquidity premia.

    Adam, As I have have said many times, I envision the market being subsidized, so your worry about not enough trading is a moot point. And the fact that no such market exists now is a point in it’s favor. Just to reiterate what I said earlier, if there was a strong demand to hedge NGDP, such a market would presumably already exist.

    If you basic argument is that the policy would not work because there would be large and time-varying risk premia, then that’s fine. I fully understand that argument. I cannot logically refute it, but I do think such a problem is very unlikely for a 12 month NGDP contract. Obviously in the real world it is unlikely that the policy would be tried without first being tested. I have no trouble with a horse race between the NGDP futures and the Fed’s internal forecasting unit. If the Fed wins, go with their forecasting unit. Right now, however, they are ignoring their own internal forecasts, so we have the worst of both worlds.

  93. Gravatar of Adam P Adam P
    3. May 2009 at 05:44

    Scott, says: ” don’t care if everyone has the same view that monetary policy is right on course. I presume they wouldn’t have that view unless it was. And if the money supply was below what they thought it would be, then the next day they would notice and trade.”

    Still not reading what was actually said. I may well believe that monetary policy is not on course but if I believe the market will eventually get it on course I don’t have any reason to trade.

    I don’t trade based on my view of whether or not monetarty policy is CURRENTLY on course. I trade based on my view of whether or not the NGDP growth target will ULTIMATELY be hit.

    Also, please read my comment responding to 123 (at 4:08). This gives the argement in more detail (before I was arguing in the context of assuming a small risk premium). See if you still think you’ve addressed the issue.

  94. Gravatar of Adam P Adam P
    3. May 2009 at 05:54

    Oh, and Scott. Elementary asset pricing theory says that an asset has a high risk premium if it’s payoff has a high correlation with aggregate consumption growth. You say the NGDP future will have a low risk premium. So:

    You think that NGDP growth has a LOW correlation with aggregate consumption growth???

    And perhaps you’d like to ask John Cochrane if he thinks risk premia are time varying or not.

  95. Gravatar of Adam P Adam P
    3. May 2009 at 06:01

    And another thing, again you’re not really reading the arguments. With respect to the negative real rate problem you said: “There is no serious economist in the world that thinks a yen dollar rate of 1000 would have been ineffective in moving Japan out of the liquidity trap…”

    I never said that either, I was always careful to say that there did exist some path for monetary policy that breaks the trap, however since that path my be very unpleasant agents might a priori doubt it will be followed. That doubt is what allows the trap to happen in the first place and your futures targeting idea doesn’t help anything in this respect.

    Furthermore, perhaps the unpleasentness of that sort of policy response explains why Japan didn’t actually do it or why Bernanke is a bit hesitant today to say, buy up every asset in the entire market (as you suggested would work on the Canadian blog).

  96. Gravatar of Adam P Adam P
    3. May 2009 at 06:14

    I should also add, just to pre-empt another silly response from you, that getting a yen dollar exchange rate of 1000 would require a commitment to a PATH for the money supply and NOT just a large supply of yen today. Why you ask? Because of efficient markets.

    If the BOJ supplies a huge amount of yen today but everyone expects that this supply is not permanent (or at least long lasting) then what happens? Well, IF the yen dollar rate hits 1000 then everyone knows that it will soon be back to 100 and they all fall over each rushing to be limit long yen. The result? Yen dollar gets nowhere near 1000 unless the large supply is believed to be permanent (or at least to last much longer than the point of just breaking the trap).

    So, this still has the same need to commit to future, unpleasant, actions and thus we have the usual problem.

  97. Gravatar of JKH JKH
    3. May 2009 at 06:25

    “It’s those differences of opinion that account for trading. If there are no differences of opinion then trading doesn’t occur, but that is appropriate because it means the current money supply is expected to produce on target growth.”

    The first part is what I’ve been saying directly and unambiguously from the start. The final inference is the obvious converse, which I’ve also stated as an inference as well. You are finally saying something in clear agreement with what I’ve said all along.

  98. Gravatar of 123 123
    3. May 2009 at 10:44

    “A Treasury-linked contract would be better than nothing, but far inferior to a NGDP futures contract.”
    Just like issuance of TIPS has stimulated the development of inflation swaps and has facilitated the monitoring of inflation expectations, I believe the issuance of NGDP linked treasures would stimulate the trade in NGDP swaps. Scott, your proposal to subsidise NGDP swap market lacks political attractiveness, but if treasury issues TGRS (Treasury GDP Related Securities) the subsidy would be hidden in the TGRS liquidity discount.

  99. Gravatar of ssumner ssumner
    4. May 2009 at 05:12

    I should probably just do a post on this soon, as I am getting responses in multiple posts:

    Adam, This is exactly my point. Some people will expect think the Fed to hit its target, and they won’t trade. Some will expect NGDP to will end up above target, and they will buy NGDP futures, some think it will end up below target, and they will sell. That’s how all markets work. The differences of opinion are essentially differences in the views regarding the future time path of velocity.

    Adam#2, Volatility is also important, and I believe it would be relatively low. In addition, if there was a strong demand to hedge consumption volatility with NGDP futures, why don’t they exist? Furthermore, the consumption angle you mention would, even if you are right, merely create a risk premium, not a time-varying one.

    Adam#3 I don’t think any monetary base path generating 5% expected NGDP growth is “unpleasant.” But if it is the wrong target, then we should use futures to target the CPI, or whatever policymakers think is the right target. I happen to think NGDP is more stabilizing than the CPI.

    AdamP#4, No commitment to future policy is necessary. If the BOJ is selling unlimited yen at 1000 to the dollar, who in their right mind would buy them from commercial banks at 100 or 500 to the dollar? The BOJ makes the market price. I think this is an important point, and something many people don’t get when they worry about whether the Fed would really be able to stabilize NGDP expectations.

    JKH, I didn’t see your point as being at all clear, especially since you seemed to argue it somehow went against futures targeting. In fact, it doesn’t create any problems at all for futures targeting if you agree with the interpretation that I just put forward. My market is like any others. Those happy with the current price don’t trade.

    123, I believe that subsidies would create no political problem at all. They would merely be higher than normal interest payments on margin accounts, and would be a minor issue compared to the cost of interest on ERs, which the public hasn’t even noticed. The public pays almost no attention to this stuff.

  100. Gravatar of Adam P Adam P
    4. May 2009 at 05:24

    Scott: “I don’t think any monetary base path generating 5% expected NGDP growth is “unpleasant.”

    Still not understanding Scott, really what’s so hard.

    Scott: “No commitment to future policy is necessary. If the BOJ is selling unlimited yen at 1000 to the dollar, who in their right mind would buy them from commercial banks at 100 or 500 to the dollar?”

    Yes, you’re quite correct my mistake. I was thinking Krugman not Svennson.

  101. Gravatar of Adam P Adam P
    4. May 2009 at 05:24

    That should have been “really what’s so hard?”

  102. Gravatar of Adam P Adam P
    4. May 2009 at 05:46

    To perhaps make things clearer let me offer a really stark hypothetical:

    Suppose that the BOJ is targeting 5% NGDP growth. Futher suppose that they find themself in situation with exactly two possibilities:

    1) The yen dollar exchange rate is greater than or equal to 1000. In this case the target is hit exactly.

    2) The yen dollar rate is less thann 1000. In this case NGDP growth will be -5%.

    Finally assume everyone knows this.

    Now, your claim is that IF expected NGDP growth can be maintained at 5% then this stabilizes aggregate demand. Let’s assume this claim is correct.

    The question is, in the example above do the japanese people expect 5% NGDP growth? Well, they do if, and only if, they believe the BOJ will set a yen dollar exchange rate of 1000 or more. Will they believe this? After all, isn’t a yen dollar rate of 1000 somewhat unpleasant for the japanese?

    Now, in the example the alternative outcome for the BOJ was so bad that yes they’d be crazy not to do the extreme devaluation but the point is that your plan still requires the central bank to commit to potentially unpleasant things in order stabilize aggregate demand. In this sense it is not an improvement on any other plan.

  103. Gravatar of JKH JKH
    4. May 2009 at 06:20

    “you seemed to argue it somehow went against futures targeting”

    The point was a clear, stand alone observation on the dynamics of the proposed market, with which you now clearly agree. I’ve never argued that this point itself went against futures targeting. I may have argued some other completely different point about NGDP policy itself, perhaps in comparison to interest rate targeting, but not this one, which has only to do with the architecture of your proposal. The opposite is exactly the case, in fact, as seems to be the norm more than not in these exchanges. In fact I said unambiguously on Nick’s blog that I quite liked the general design of the futures contract with OMO – after the onion had been peeled off a bit and the proposed mechanics for the intervention finally revealed. And you responded to that comment! You don’t read the comments closely enough, and you seem always to be moving the facts around to avoid admitting errors or false charges along the way.

  104. Gravatar of Adam P Adam P
    4. May 2009 at 06:44

    JKH, I will say one thing in Scott’s defense that occurred to me a couple of comments up when I had to say “my mistake” for what was a rather dumb error on my part. One of the reasons it was so easy for me to admit the mistake is because I’m doing this anonymously. Scott isn’t.

    On the other hand, maybe that’s just another reason for Scott not to be condescending to his commenters.

  105. Gravatar of Niklas Blanchard Niklas Blanchard
    4. May 2009 at 10:01

    If you didn’t catch this, I thought it might be interesting regarding your question about Tyler Cowen:

    Q. I’m curious about the time commitment of your blog. In a typical day, how much time do you spend on the following:

    A. Writing on MR — One hour?
    Thinking about what to write — Close to zero, or during downtime.
    Reading comments on MR — 20 minutes? If I’m traveling I don’t read them at all.
    Reading other blogs — 40 minutes?
    Reading and writing blog-related emails — I get quite a few, I’m not sure what it amounts to in terms of time. I do try to respond to almost everybody.
    Looking at traffic statistics — Zero, this is counterproductive.

    The big time cost is reading but of course I would be doing that anyway.

    http://wehrintheworld.blogspot.com/2009/05/q-with-tyler-cowen.html

  106. Gravatar of ssumner ssumner
    5. May 2009 at 05:39

    Adam, There are two issues here:

    1. Is NGDP targeting perceived as being such a bad idea that the public expects (maybe correctly) that the government will abandon it in the near future?

    2. Is the current (today’s) NGDP peg credible?

    In my system, each day is a new contract. As long as they peg that contract, it is credible FOR THAT DAY, just like the 1000 yen example. And just as in that case, if they think the government will realize that for political reasons it is a big mistake, and expect the peg to change in the future, that may occur with NGDP targeting as well. It goes without saying that one cannot commit from now until the end of time, or maybe even until next week. But one can commit today if you buy and sell unlimited contracts today. Now I know you don’t accept the risk premium argument, but putting that aside, they can force the public to expect 5% NGDP growth over the next 12 months, but cannot force them to expect that growth rate from one month from now, to 13 months from now. Those contracts haven’t been traded yet, and the government might decide never to trade them.

    JKH, Sorry, that won’t work. You were clearly arguing the price peg would fail, in the sense that a currency peg could fail. And you were wrong. What is true is that the actual future NGDP might come in more or less than expected, but that is a different issue. Even if many individual investors think NGDP will not come in at 5%, the NGDP contract price will stay at 5% as long as the government pegs it. Here is what you said:

    JKH
    30. April 2009 at 05:28 (#)

    “There would be no private sector participation in such an NGDP futures market at all, unless such participants perceived a risk that the Fed’s NGDP price peg would fail. Otherwise, the operation of such a market is a contradiction – there would be no need for private sector hedging or desire for speculation. The only reason there would be a two sided market in NGDP futures is if participants believed there was a risk that the NGDP target wasn’t credible.

    An analogy is (previous) hot money flows into China, based on the expectation that China would break the RMB peg and revalue.

    Another analogy is changes in the yield curve according to expectations for the fed funds rate.”

    Those analogies show that you don’t understand the idea.

    Thanks Niklas.

  107. Gravatar of Efficient Markets, as they stand. « Rortybomb Efficient Markets, as they stand. « Rortybomb
    5. May 2009 at 19:57

    [...] The Money Illusion: I think about that failure in terms of efficient markets theory…. [...]

  108. Gravatar of Qsbubba2 Qsbubba2
    15. December 2009 at 16:01

    I think its interesting that you said predictions were not accurate and that no one can tell whats coming. I have only read the preface of a book “How to survive hyperinflation” because it scared me. It was written around 1989 and it listed the general ways that EVERY nation that has gone into hyperinflation got there. These are all things that the USA has done since just before ’89. I knew before I read the preface we were in trouble but i was hoping for a ‘depression’ to reset our problems. Countries that have a depression usually still have the same $ and therefore the same or similar economic foundations. In ALL hyperinflation ntns i have read about there was a total destruction of the old economy & $ system followed by a nightmarishly tyranical govt/ dictator. We are completing the final steps in the preface. There were financial crises followed by major govt spending. Shortly thereafter came inflationdue to all outsiders devaluing the $$ then hyperinflation. I have read of a King’s cutiing their silver coin by 50% to fund a war and within6 months there was a total economic collapse(this was more tban 900 years ago. Inflation and its devastation is not a new idea its old and its stupid. We are going to fall its only how far and when… What you were complaining about(writer of blog) wasnt really that they couldnt predict what would happen because THEY DID…you wanted them to tell you WHEN. Which no one can predict exact dates for the future…thats why they call it the future instead of the past. As long as we move forward in de-industrialising our nation we will keep digging a deeper and deeper grave. Almost all republics have risen from the ashes but none were as great as they were before the fall. If we keeping dighing maybe we wont rise again either!!
    Most people i talk to that are educated in economics are so brainwashed by $$ they can’t understand basic historical principles and logic. We are giving all countries digital/paper $$ for hard goods Just like before and during all hyperinflation economies!!! Our $$ is not backed by anything stable… Not gold not houses not oil…you cannot print/digitize $$ for house that you say is worth $25k in 1970
    and then do the same thing on the same house in 2008 & say its worth $750k without eventually devaluing $$ by 3000%. Which the housing market supply and demand has remained fairly stable…for every new family in USA we have built a new home =STATUS QUO(more or less but anything else is fantasy or symantics)!!

Leave a Reply