The Fed doesn’t have the luxury of waiting for the perfect DSGE model; they must do policy right now

OGT sent me a recent Stephen Williamson post on Christina Romer’s call for NGDP targeting.  Williamson is highly critical, but doesn’t seem to fully understand the distinction between policy instruments and policy goals:

NGDP targeting does not do anything other than specify the Fed’s ultimate goals.

Goals are macroeconomic outcomes that the central banks would like to see.  Instruments are actions or statements by the central bank that change the current supply or demand for base money.  Adoption of NGDPLT is not just a policy goal, it is an act that changes the current demand for base money.  And it’s not just a policy tool; it’s by far the Fed’s most powerful tool.  The Fed steers the nominal economy by sending out signals regarding the future path of the supply and demand for base money.  Even changes in the fed funds target work this way—no one seriously believes the huge stock market responses to the January 2001, September 2007 and December 2007 Fed announcements occurred because the target was  set 1/8% away from market expectations.  Rather they changed expectations of the future policy trajectory.  It’s always about the expected future path of policy.

But Williamson is right about his main point; Romer doesn’t present a rigorous model that would justify NGDPLT.  Perhaps such a model doesn’t exist.  But here’s the problem, no one else has a rigorous model that would justify any monetary policy, at least not a model that passes the laugh test.  Just imagine if Steve Williamson walked into the FOMC with his favorite GE model and said; “Here’s the correct model of the economy; this tells you how you should conduct monetary policy.”  People would literally fall on the floor laughing.

In the comment section someone mentioned Evan Soltas’s recent defense of NGDPLT, and Williamson responded:

http://tinyurl.com/7k8db4f

You realize that was written by a high school student?

“Perhaps not all of these arguments are wrapped up in one tight analytical model, but these general concepts should not be difficult to model in chunks.”

How do I model in chunks?

Charming.  I’d recommend Williamson read Milton Friedman, if he wants to see partial equilibrium analysis that blows away 99.9% of the GE stuff churned out by our grad schools today.

The hard truth is that the Fed must do monetary policy.  We will never find an even half way decent DSGE model of the economy, so they need to rely on pragmatic arguments.  If you read the minutes of the Fed meetings, you’ll see exactly the same sort of non-technical pragmatic arguments that Evan used.  The debate at the Fed almost never rises above the sort of simple, natural rate AS/AD model used in intro econ classes.

Here’s the type of argument that economists find persuasive:

This may not be news to you, but it was to me. In this speech, Narayana Kocherlakota shows us how the labor market behavior in Sweden, following the early-1990s financial crisis that occurred there, looks much like what has been happening recently in the United States. This is consistent with this post, where I looked at some Canadian labor market data. Canada sailed through the financial crisis with essentially no problems in its banking sector, and the recent behavior of the labor market (or rather, labour market) in Canada looks quite different from the US. Sweden had a financial crisis in the early 1990s, and it’s labor market behaved subsequently like the US labor market is behaving now.

You may or may not be persuaded by that particular argument (I’m not), but the point is that it represents exactly the sort of argument that every macroeconomist from Krugman to Cochrane uses.  And they use them because they know other macroeconomists are far more persuaded by this sort of reasoning, then if they say “let me show you my newest mathematical GE model; it tells us what the Fed should be doing.”  And I’m pretty sure that Williamson knows this, as I pulled the argument from his latest blog post.

PS.  I was puzzled by this statement:

Romer and the FOMC are on the same page on this one, but I don’t think QE does anything at all (again, see the last link above). At best, QE can signal future intentions of the Fed with regard to the policy rate (and thus move asset prices), but the Fed can do the same thing with “forward guidance,” i.e. announcements about the future path for the policy rate.

Oh really?  What if the Fed announces zero interest rates for the next 20 years?  Is that easy money or tight money?  Williamson calls himself a “new monetarist,” but I can’t help wondering what Friedman would think of an economist claiming that the future path of nominal interest rates describes the stance of monetary policy.


Tags:

 
 
 

47 Responses to “The Fed doesn’t have the luxury of waiting for the perfect DSGE model; they must do policy right now”

  1. Gravatar of D.Gibson D.Gibson
    12. June 2012 at 07:53

    There is a lot of ignorance out there, when it comes to NGDPLT. I encourage you to improve the Wikipedia entry for NGDPLT. It is the world’s primary source for a quick explanation of everything and the NGDPLT section is just a few sentences in a larger article. I think a new, dedicated article would be better.

  2. Gravatar of Major_Freedom Major_Freedom
    12. June 2012 at 08:27

    ssumner:

    The Fed doesn’t have the luxury of waiting for the perfect DSGE model; they must do policy right now

    It’s funny how “policy” is now being used to refer to “money devaluation.”

    “Policy” has an aura of sophistication and improvement.

    “Money devaluation” is more accurate, but it is politically less popular. The Joe Schmoes would know what is going on. They would know their money and incomes are being drained of value. With “policy”, they are more easily duped.

    Orwell would be proud.

    Re: The Friedman quote…

    Didn’t Friedman say low interest rates are “generally” a sign that money has been tight? That doesn’t mean “always.”

    In other words, it could mean that the low interest rates post-2008 are NOT due to “tight money”, and is just a continuation of a decades long bull market in bonds. It could be one of those times that are not a “generally” time.

    If you look at this chart, you will see that occasionally, interest rates did not go up when NGDP went up, and did not go down when NGDP went down.

    Late 1980s to the 2000s saw NGDP growth relatively stable, and yet interest rates continued to fall.

    Post-2010 has seen NGDP growth relatively stable, and yet interest rates continued to fall.

  3. Gravatar of Mike Sax Mike Sax
    12. June 2012 at 08:30

    Interesting. As I’ve suggested before I don’t really know which horse to back in this argument over models and “microfoundations.”

    “the sort of argument that every macroeconomist from Krugman to Cochrane uses. And they use them because they know other macroeconomists are far more persuaded by this sort of reasoning, then if they say “let me show you my newest mathematical GE model; it tells us what the Fed should be doing.”

    In a way I find this a relief, because it suggests that macro people don’t think as differently as the layman as you’d might think. Clearly these are the kinds of arguments I find more persausive too.

  4. Gravatar of Steve Williamson Steve Williamson
    12. June 2012 at 08:31

    1. “Williamson is highly critical, but doesn’t seem to fully understand the distinction between policy instruments and policy goals…”

    I think you need to tell us what the operating procedure is that you are recommending. In your writings you seem to be saying that the Fed came somehow wish to hit their NGDP target, and it will be so. Under a NGDP target, how does the Fed use the information it has on actual NGDP to issue a directive to the open market desk, and what exactly is the nature of that directive?

    2. “I’d recommend Williamson read Milton Friedman, if he wants to see partial equilibrium analysis that blows away 99.9% of the GE stuff churned out by our grad schools today.”

    Actually, I think I’ve read about 90% of Friedman. Friedman was brilliant, but unfortunately limited by his technical knowledge. Sometimes he was “right,” for example with the Friedman rule (zero nominal interest rate in all states of the world). That’s something you can flesh out in many standard monetary models, though we think there is something wrong with the Friedman rule as practical policy. Sometimes Friedman was wrong, for example with the 100% reserve requirement. He didn’t understand financial intermediation. You shouldn’t take Friedman as an example of why we shouldn’t do math. Friedman could have been much better if he had known as much math as Lucas does.

    3. ” If you read the minutes of the Fed meetings, you’ll see exactly the same sort of non-technical pragmatic arguments that Evan used. The debate at the Fed almost never rises above the sort of simple, natural rate AS/AD model used in intro econ classes.”

    Have you ever talked to anyone who has been to an FOMC meeting? Bernanke, Kocherlakota, Lacker, Plosser, Evans, Rosengren, Bullard, at least, are sophisticated economists with excellent technical skills. Natural rate AS/AD models would actually get the guffaws in that room.

    4. “Here’s the type of argument that economists find persuasive:”

    When you say “economists,” I think you mean the readers and writers of blogs. You should understand that not everyone in the economics profession takes the blogosphere seriously. The quote from my blog post is just a piece of information which might suggest something, but it’s not serious economics. If you show up at the SED meetings in Cyprus next week, you’ll actually see “economists” doing serious work.

    Steve

  5. Gravatar of Mike Sax Mike Sax
    12. June 2012 at 08:33

    I got so excited after understaind Lars Sumner Phillips rule or whatever that I went and applied some-very basic-modeling myself.

    http://diaryofarepublicanhater.blogspot.com/2012/06/rush-limbaughs-shocking-ignorance-about.html

  6. Gravatar of Mike Sax Mike Sax
    12. June 2012 at 08:38

    “”Money devaluation” is more accurate, but it is politically less popular. The Joe Schmoes would know what is going on. They would know their money and incomes are being drained of value. With “policy”, they are more easily duped.”

    Yep. The “Joe Schmoes” of the world are outraged by any inflation. However, Herman Hoppe and Major Freedom’s agenda to take away suffrage is of no concern to this same Joe Schmoe.

  7. Gravatar of dwb dwb
    12. June 2012 at 08:48

    thanks for commenting on SW’s post. It’s frustrating: QE does not do anything except perhaps a wealth effect and provide forward guidance. oh is that all . well, that proves it then!

    Some of the comments over there made me want to throw a brick at the screen.

  8. Gravatar of Becky Hargrove Becky Hargrove
    12. June 2012 at 08:49

    Mike,
    One way I think about it: NGDP as the pool of already existing possibilities for the marketplace and those who participate in it. Whereas, the limitations of credit thinking and interest rate targeting are but a portion of that marketplace, which attempt to define the whole thing on their own terms.

  9. Gravatar of Major_Freedom Major_Freedom
    12. June 2012 at 08:52

    Mike Sax:

    Yep. The “Joe Schmoes” of the world are outraged by any inflation. However, Herman Hoppe and Major Freedom’s agenda to take away suffrage is of no concern to this same Joe Schmoe.

    If by “taking away suffrage” you mean allowing individuals the freedom to not be ruled by democratic majorities, and to live in their own communities, without forcing you into their communities, then sure, the agenda is to “take suffrage away” from you.

    It is very much like “taking away” a parasite’s host, and then listening to the parasite whine that his democratic rights are being violated.

  10. Gravatar of Mike Sax Mike Sax
    12. June 2012 at 09:03

    Intersting way of looking at it Becky

  11. Gravatar of dwb dwb
    12. June 2012 at 09:16

    When you say “economists,” I think you mean the readers and writers of blogs. You should understand that not everyone in the economics profession takes the blogosphere seriously.

    I have never convinced anyone of anything with a model. However, the english translation of that model can be convincing, or not, once the assumptions are laid bare (in english, i have always been of the cranky opinion that a technical PhD was useless without communication skills). As far as an actual model, it does not prove anything since garbage in=garbage out.

    Of course the fact that FOMC members feel compelled to respond to whats in the blogosphere should signal that its more “serious” than many people think (recall Bullard responding to Duy a while back). 3297 heads are better than 1.

  12. Gravatar of Lars Christensen Lars Christensen
    12. June 2012 at 09:24

    Scott, this whole story with Swedish unemployment in the 1990s is very odd. What did they expect? The Riksbank tigthened monetary policy insanely (overnight rates was hiked to 500%!) and the Swedish financial sector more or less collapsed as a consequence. This have nothing to do with supply side factors – Riksbanken caused NGDP to collapse and as a consequence unemployment spiked.

    And then he and Kocherlakota wonder why US unemployment is showing the same bahaviour. Well, maybe that is because US monetary policy is insanely tight? Did the “New Monetarist” think about that?

    And then he argues that Canada did not see the same kind of rise in unemployment – did Williamson care to have a look at NGDP in Canada? BoC eased monetary policy to avoid a collapse in NGDP and as a consequence unemployment is not “Swedish”.

    By the way there has been a significant shift in Sweden into what the Swedes calls “open unemployment” from labour market programmes where unemployment earlier used to be hidden away. This have nothing to do with Sweden’s 1990s crisis but happened to coincide with the permanent increase in Swedish unemployment.

    Finally David Eagle has in a couple of his papers shown that there historically have been a very close correlation between NGDP growth and unemployment. In in post-recession periods where NGDP has returned fast to the pre-rececssion trend unemployment has also come down fast. Obviously NGDP is stuck well below the pre-crisis trend and therefore it should hardly be an surprise if you have read Eagle that US unemployment remains elevated.

  13. Gravatar of ssumner ssumner
    12. June 2012 at 09:56

    D. Gibson, I don’t have time–perhaps one of my commenters could improve it.

    Steve, Thanks for replying. You said;

    “I think you need to tell us what the operating procedure is that you are recommending.”

    Ever since the late 1980s I’ve been publishing papers advocating the Fed rely in NGDP futures (or CPI futures.) I’d have them adjust the base to keep NGDP expectations growing along a steady 5% growth track, level targeting.

    I have a paper in the Berkeley journal “Contributions to Macroeconomics” in 2006 that lays out the general approach. Interestingly, Narayana Kocherlakota (and economist you seem to like) just endorsed my proposal last Friday (as I discuss in my new post.) John Cochrane endorsed it a couple years ago, after I explained the idea to him. So it looks like we low tech guys are sometimes ahead of the curve.

    You said;

    “You shouldn’t take Friedman as an example of why we shouldn’t do math. Friedman could have been much better if he had known as much math as Lucas does.”

    I didn’t say you shouldn’t do math, I’ve also published mathenmatical models. The point was that math isn’t the be-all-and-end-all of macroeconomic analysis. Friedman’s views of business cycles (as for instance the Monetary History–still the most influential applied macro book ever written) have held up far better than Lucas’s monetary misperceptions model. I’m a huge fan of Lucas, but mostly for rational expectations/Lucas critique reasons; he wasn’t able to come up with a persuasive macro model. BTW, Lucas thought The Monetary History was the best empirical study of the effects of money, despite it being very low tech.

    Friedman was an extremely intelligent guy, he learned whatever math he felt he needed to get this points across.

    You said;

    “Have you ever talked to anyone who has been to an FOMC meeting? Bernanke, Kocherlakota, Lacker, Plosser, Evans, Rosengren, Bullard, at least, are sophisticated economists with excellent technical skills. Natural rate AS/AD models would actually get the guffaws in that room.”

    This is a complete non-sequitur, which has no bearing on the point you just quoted. I was describing the sorts of arguments used in the FOMC, it’s 99% applied AS/AD. Sure these guys know much more than that, I never claimed otherwise. So do I. But I read the minutes, and they basically talk about demand shocks, supply shocks and changes in the natural rate of output/unemployment. That’s how they make their decisions.

    Regarding “serious” economics, I’d suggest you read Deirdre McCloskey on methodology. The myth that there are two types of economics, serious and popular, is very attractive to many economists, but doesn’t hold any water. Economics is 100% about persuasion, sometimes highly technical arguments are helpful, sometimes they aren’t. I see even the most highly technical economists being swayed by trivial stylized facts all the time. It’s how things work in the real world.

    Ultimately people like Plosser and Kocherlakota are judged not on their scholarly papers, but rather on whether they can make cogent arguments for their policy recommendations.

  14. Gravatar of ssumner ssumner
    12. June 2012 at 10:02

    dwb, I agree.

    Lars, Williamson said that the direction of causation between macro economy and financial crises isn’t clear. I agree, but I suspect that there’s far more causation going from declining NGDP to financial distress than most economists believe.

    I completely agree about the close correlation between NGDP and employment.

  15. Gravatar of Tim Tim
    12. June 2012 at 10:05

    Hi scott
    I know i have to stop writing off topic comments but i want to ask this now otherwise ill forget by the time its relevant! The other day i asked why you thought that the Australian dollar appreciated when the RBA cut rates. How about this: if an inflation targeting central bank is targeting the forcast, and inflation is stable, then expansionary monetary policy to keep inflation at that trend won’t change the future price level and hence the exchange rate wont depreciate. However, if the fed engaged in a higher inflation target (level targeting) it would signify a new trend compared to the 0-1% trend since 2008, and hence a higher expected price level, and the currency would depreciate. So it all depends on whether previous and current monetary policy has kept inflation at trend..

  16. Gravatar of Steve Williamson Steve Williamson
    12. June 2012 at 10:45

    “I’d have them adjust the base to keep NGDP expectations growing along a steady 5% growth track, level targeting.”

    That’s not an operating procedure. The people who are doing open market operations at the New York Fed can’t see “NGDP expectations.”

    “I didn’t say you shouldn’t do math, I’ve also published mathenmatical models.”

    That seems to be the flavor of what you’re saying, though. The gist of your argument is that the real world does not need formal modeling, and you’re claiming that in practice that’s what happens. Neither is correct. We do the formal modeling for a reason. It helps to make our ideas precise, and sometimes surprises us. An FOMC discussion is informed by a lot of formal, hard work, that is done by the people who work in the Fed system, and that goes into briefings before the meeting takes place. The participants carry that formal analysis in their heads when they are in the room, but of course they have to communicate that information in terms that most people in the room will understand. That said, I’m assured by the people I know who actually attend this things currently, that the level of discussion is very high – much above AS/AD, thank you.

    I like McCloskey’s arguments about persuasion. I learned a lot from Don/Deirdre when he/she was my colleague. But I think McCloskey would include the formal modeling as part of the persuasion. I certainly do. If you want to persuade me, and I think most mainstream economists, you’re going to have to show how this works formally. Don’t give me this bullshit about how models are useless.

  17. Gravatar of Max Max
    12. June 2012 at 11:09

    “What if the Fed announces zero interest rates for the next 20 years? Is that easy money or tight money?”

    Neither. But it’s not a meaningless statement! The path of interest rates tells you how much you should pay for government bonds (or interest rate futures). The markets care about interest rates, independent of the macro effects.

  18. Gravatar of Major_Freedom Major_Freedom
    12. June 2012 at 11:46

    ssumner:

    I suspect that there’s far more causation going from declining NGDP to financial distress than most economists believe.

    Why can’t financial distress cause declining NGDP?

    A huge chunk of the aggregate money supply and hence aggregate spending is composed of fiduciary bank credit. What if there is financial distress, and the growth rate of bank credit plummets, or turns negative, and there are defaults on existing credit, both of which can lead to a decline in aggregate money supply, and thus aggregate spending?

    If banks in NY are exposed to bad mortgage debt, how will increasing spending, say in the military industrial complex, solve this problem?

    Focusing on aggregates is far too sloppy an approach. So many details and nuances are overlooked.

  19. Gravatar of Major_Freedom Major_Freedom
    12. June 2012 at 12:07

    Lars Christensen:

    this whole story with Swedish unemployment in the 1990s is very odd. What did they expect? The Riksbank tigthened monetary policy insanely (overnight rates was hiked to 500%!) and the Swedish financial sector more or less collapsed as a consequence. This have nothing to do with supply side factors – Riksbanken caused NGDP to collapse and as a consequence unemployment spiked.

    You can’t say unemployment spiked “as a consequence” of the fall in NGDP by looking at the correlation alone. For the theory that NGDP fell “as a consequence” of unemployment spiking is also entirely consistent with the data.

    Also, the theory that the Riksbank inflated during the 1980s up until the early 1990s, which caused labor to go into unsustainable lines of production, the errors of which were revealed after the Riksbank raised interest rates and reduced inflation, is also entirely consistent with the data.

  20. Gravatar of Adam Adam
    12. June 2012 at 12:08

    Professor Williamson – I think you should know that you come off as highly condensending in this exchange, as well in recent exchanges with Miles Kimball as well. Perhaps that’s because you honestly feel they lack your technical skill or insight.

    But if you think engaging them in this way is going to make you more persuasive, you’re wrong.

  21. Gravatar of Wonks Anonymous Wonks Anonymous
    12. June 2012 at 13:10

    It seems Williamson hails from the concrete steppes. His fellow Canuck has a tailor-made explanation of NGDP targeting:
    http://worthwhile.typepad.com/worthwhile_canadian_initi/2011/10/engdp-level-path-targeting-for-the-people-of-the-concrete-steppes-/
    It’s true that the NYFed does not have direct access to a crystal ball of NGDP. But they could set up a futures market (as Scott has proposed). They could also rely on an internal forecast. They might be bad at forecasting, in which case a futures market sounds much better.

  22. Gravatar of johnleemk johnleemk
    12. June 2012 at 13:39

    Wonks Anonymous, your link is broken. This for some reason works: http://worthwhile.typepad.com/worthwhile_canadian_initi/2011/10/engdp-level-path-targeting-for-the-people-of-the-concrete-steppes-/comments/page/2/

  23. Gravatar of OGT OGT
    12. June 2012 at 13:40

    Scott- Glad you turned this into a post.

    How valuable is internal consistency with poor assumptions? The Catholic Church has 2000 years of scholarship on that point that isn’t terribly encouraging.

  24. Gravatar of Mike Sax Mike Sax
    12. June 2012 at 14:53

    Sumner-Willamson dust up over models

    http://diaryofarepublicanhater.blogspot.com/2012/06/scott-sumner-stephen-williamson-dust-up.html

  25. Gravatar of dwb dwb
    12. June 2012 at 17:12

    this whole story with Swedish unemployment in the 1990s is very odd.

    i’d like to know enough to bat down Kocherlakota’s points, the danger there is that it just sounds plausible enough to be true. I get that money was tight in the early 90s (along with a bunch of other changes I read about like changes toto the mortgage interest deduction). I am not sure i understand why it then only went back down to 6%, and even then took so long (real estate crash?). I looked through some academic papers including some from the Riksbank and did not come away with a feel that this was well understood. It would make more sense to me that the 2-3% from the pre-1990 period did not account for all the unemployed, but thats what the official stats say.

  26. Gravatar of ssumner ssumner
    12. June 2012 at 17:33

    Tim, That’s part of it, but the bottom line is that any market response should reflect the unexpected part of the announcement.

    Steve, Regarding instructions to the bond desk, I’d have them buy bonds every time someone sold NGDP futures to the Fed (at a price equal to the Fed’s NGDP goal), and vice versa. I’m not sure the ratio, perhaps 100 to 1. The market would essentially be setting the monetary base at the level expected to produce on-target NGDP growth. I don’t trust the Fed to set the money supply or the fed funds rate, I want them to set out the policy goals, and have the market actually implement policy.

    There are two types of models. One type tries to show the structure of the economy, and which policy target is optimal given that structure and given the assumed social welfare function. Those models are not convincing, to put it mildly. There is no consensus on any sort of monetary policy except perhaps “keep inflation fairly low and stable on average, and try to keep employment near the natural rate if possible.” Beyond that it’s all open to debate, and we are not closing in on any sort of consensus, indeed we are spiralling ever further apart. That’s why contrary to what you claim, FOMC meetings take place at a very elementary level–I still insist it’s almost all AS/AD stuff, regardless of how much these guys actually know (and many like Fisher are complete fools, if his public comments are any indication.) The minutes are released to the press. I’ve read them. I don’t know if you have, but it’s a very sobering experience to read the discussion.

    The other type of model is aimed at giving the Fed information on how to set the policy instrument, given a policy goal has been set. I don’t think much of those models either, and instead agree with Kocherlakota that we should let the markets tell us where the money supply should be set. I’d abolish the economic research units at the various Feds, and rely solely on market forecasts.

    Glad we agree on McCloskey. I think models are very useful, I just have a more expansive definition of “models” than most macroeconomists. I think my blog is full of models, and the entire blog as a whole is a sort of mega-model, albeit admittedly a sprawling mess in organization. I hope to soon write a book that will whip it into some sort of shape. The book will contain a model, but no math.

    Max, I agree.

    Wonks, I should have mentioned the Svensson approach, if there is no futures market, then you are forced to rely on internal forecasts.

    OGT, Yes, Isn’t there an old saying about wanting to be be approximately right rather than precisely wrong?

  27. Gravatar of Yichuan Wang Yichuan Wang
    12. June 2012 at 18:50

    Steve,

    The NGDP futures strategy is how Scott measures NGDP expectations (http://www.themoneyillusion.com/?p=11550). The Fed could use the price of the NGDP future to see what the market expectation of NGDP growth is, much as the TIPS spread can be used to gauge the market expectation of inflation expectations. With an NGDP futures market, the folks conducting OMO’s CAN see NGDP expectations.

    When it comes to operating procedure, what you seem to be missing is that the Federal Reserve has a lot of ammunition at its disposal. It can depreciate the currency, it can monetize parts of the debt, or as Scott Sumner has frequently pointed out, it can lower IOR to try to get the reserves out of the banks. The fact that these tools \have the ability to drive the value of the currency to zero means that the Fed can have a credible commitment to a nominal policy target, such as an NGDP. Perhaps this won’t send real growth skyrocketing as well, but a stabilization of THE key nominal aggregate when it comes to debt contracts and safe assets would go a long ways towards stabilizing long term results.

    And these results are confirmed by “chunks” of models. David Beckworth’s work on the safe assets story has shown how NGDP shocks are a key factor on the production of safe assets. Formal analyses of shadow banking then explain how this safe asset problem can spread throughout the financial system. A recent Kiyotaki and Moore model (http://www.nber.org/papers/w17934) explains how the illiquidity of certain assets purchased by the Federal Reserve GUARANTEES traction from monetary policy (Although monetary expansion would be greatly helped by lower IOR). It should not be too difficult for the Fed to commit to a concrete policy UNTIL an NGDP target is hit. In conventional terms, it could mean a joint target of the unemployment rate or inflation rate, as Christina Romer wrote in that op-ed that you criticized.

    Another powerful property of this policy is that long term expectations can be used to power NGDP growth now. In formal terms, you can imagine a guarantee from the Fed that, in n periods, NGDP growth will be back to its 4.5% trendline. If that is true, then it would make no sense to hold so much cash/safe assets in the n-1 period, and by backwards induction, the market would act as if there’s a stable NGDP growth path in the present. This game theory argument would not function for a simple inflation target, because the current level of inflation does not result in low enough of real interest rates. Even if the market can credibly expect inflation to be 2% in the future, the lack of catch-up growth would prevent a sufficient change in the asset allocations of the current market. This is just a more formal model thinking of Nick Rowe’s “Chuck Norris beating people in rooms” model. While NGDPLT means Chuck Norris threatens to beat more people if each successive period, IT means Chuck Norris only threatens to beat one person in each period. If beating one person isn’t enough to get everybody to move in the current period, there’s no reason to expect that threat to gain any power in future periods. However, if the threat progressively grows, then expectations about increased future beatings will stimulate movement into the other room today.

    These “chunks” help make up the narrative of NGDPLT, and are likely to be building blocks for more formal models. I would hope that you do not confuse absence of evidence as evidence of absence. Simply because we have not built DSGE models doesn’t mean the formal arguments for NGDPLT don’t exist. Thus, the lack of a formal unified model should motivate us to continue studying the concept, and not to just give up and say NDGPLT is a failure because the math isn’t there yet.

    Yichuan Wang

  28. Gravatar of JoeMac JoeMac
    12. June 2012 at 18:54

    Scott,

    How would the Fed target NGDP if there don’t exist NGDP futures?

  29. Gravatar of Jason Odegaard Jason Odegaard
    12. June 2012 at 19:01

    Steve Williamson,

    Here is the blog post you are looking for, where Scott details how to create an NGDP futures market that guides the daily Treasuries trading activities of the Fed.

    http://www.themoneyillusion.com/?p=1184

  30. Gravatar of johnleemk johnleemk
    12. June 2012 at 19:14

    Yichuan Wang is a high school senior/rising freshman at UMich, according to his blog…geez. Market monetarism seems to be attracting all the economics-literate high schoolers out there; first Evan Soltas, now Yichuan Wang. Who’s next?

  31. Gravatar of Lorenzo from Downunder Lorenzo from Downunder
    13. June 2012 at 03:14

    The post, by an Australian economist, expresses graphically just how much Australia is in a different place:
    http://tomjconley.blogspot.com.au/2012/06/not-just-lucky-good-australian-economy.html

  32. Gravatar of ssumner ssumner
    13. June 2012 at 08:03

    Yichuan, Excellent comment! BTW, I will be visiting China later this summer.

    JoeMac, I’ve proposed that the Fed set up and subsidize trading in an NGDP futures market.

    Jason, Thanks for that link.

    Johnleemk, Things sure have changed from when I was in high school!

    Lorenzo, People should check out that link, there are some very interesting graphs.

  33. Gravatar of Yichuan Wang Yichuan Wang
    13. June 2012 at 08:41

    Scott,

    If you’re ever in Shanghai this summer, just shoot me an email. I’m there for almost the entire summer and I’d love to meet you.

    On a more substantive note, the recent discussion/getting blown off by Steve led me to think a bit more about forward guidance.

    First, I remember the old banter between you and Nick Rowe on how the future path of the interest rate was about 99% of the effect, but some recent econometric papers have found that the future path doesn’t exert that large of an effect. It’s ONLY about 80% (http://bit.ly/MBuxdq).

    Second, I think our skepticism about interest rate targeting only makes sense in a world in which the Fed doesn’t commit to aggressive policy. Yes, Japan committing to a 0% interest rate for a decade could just be a sign that there’s no chance for expansion, but what if the BoJ committed to that interest rate path in addition to aggressive asset purchases and further unconventional easing. The unconventional easing would lead to rising NGDP growth, and if the BoJ commits to its low interest rate path the policy could become much more credible. It would stop being a prediction of bad economic conditions and instead become a commitment to fostering higher NGDP growth.

    I wrote about these in a longer post on my blog today, and I was wondering what you thought about it.

    http://synthenomics.blogspot.com/2012/06/limits-of-forward-guidance.html

    Yichuan Wang

  34. Gravatar of Major_Freedom Major_Freedom
    13. June 2012 at 09:07

    Yichuan wrote:

    My first reaction is the classic Market Monetarist/Friedman/Bernanke retort: interest rates aren’t an indicator for the stance of monetary policy. Perhaps the Fed lowering the interest rate could be an example of loosening policy, but it would hardly qualify as loose policy. In a sense, the derivative of the interest rate time path could give information about the derivative of the policy tightness function, but the level of the interest rate tells you nothing about the tightness of policy. Low interest rates could reflect low NGDP expectation or they could reflect substantial levels of monetary easing. There’s no way to actually tell.

    You see that? Even high schoolers outclass Sumner.

    Yichuan is exactly right…on this point at least. By observing the data alone, there is no way to refute or verify a specific theory among competing, mutually exclusive theories. There has to be a different standard than past data to settle things like this.

    I get scolded for saying the same thing Yichuan says. Ah well, at least I know I’m right despite the perpetual antagonism.

    Yichuan, since you’re no fool, I highly suggest you make some of the people here rage by familiarizing yourself with praxeology and thymology, and Austrian economics.

  35. Gravatar of Jeff Jeff
    13. June 2012 at 11:27

    If you compare today’s FOMC to the FOMC of thirty years ago, there is no doubt that the level of mathematical sophistication has risen greatly. As Steve Williamson says, many of today’s committee members are accomplished modelers themselves, and the staff regularly presents results from assorted DSGE and other models that could not have been estimated or simulated with the technology of 30 years ago. I don’t want to out myself here, but I really am in a position to know this.

    And yet, if you were to tell a general audience that monetary and macroeconomic policy is better today than it was thirty years ago, I think you’d be laughed out of the room. Why is that? Could it be that the profession took a wrong turn somewhere?

  36. Gravatar of Jeff Jeff
    13. June 2012 at 11:28

    And no, Major Freedom, it’s not because the FOMC is less Austrian now than it was 30 years ago.

  37. Gravatar of Charlie Charlie
    13. June 2012 at 13:57

    Stephen has a serious internal contradiction. He says that QE can work through the expectations channel, but at the same time thinks monetary policy can’t be stimulative. He concedes that Fed policy can work through information about future policy targets, and in this post even explains that the Fed is refusing to give much forward guidance. He has a reader find the Fed quote below, where the FOMC details why it REFUSES to give forward guidance. It explains exactly why a level targeting approach would work and would be a huge policy change, even if the instructions to the operating are unchanged.

    FOMC minutes:

    “Several participants thought it would be helpful to provide more information about the economic conditions that would be likely to warrant maintaining the current target range for the federal funds rate, perhaps by providing numerical thresholds for the unemployment and inflation rates. Different opinions were expressed regarding the appropriate values of such thresholds, reflecting different assessments of the path for the federal funds rate that would likely be appropriate to foster the Committee’s longer-run goals. However, some participants worried that such thresholds would not accurately or effectively convey the Committee’s forward-looking approach to monetary policy and thus would pose difficult communications issues, or that movements in the unemployment rate, by themselves,
    would be an unreliable measure of progress toward maximum employment.”

  38. Gravatar of Yichuan Wang Yichuan Wang
    13. June 2012 at 14:03

    Major Freedom, my point about the interest rate doesn’t mean that the current low interest rate environment isn’t contractionary. The current regime is contractionary because NGDP growth is below trend. No, the interest rate in and of itself can’t define the stance of MP, but it in conjunction with other indicators can help define monetary policy.

    And truly, there’s no way to identify the stance of monetary policy outside of its objective, which market monetarists identify to be NGDP. Beckworth’s work on safe assets and other work on the problems in the shadow banking center have shown that NGDP growth can have effects on what is even considered to be money. In effect, money is regime dependent; it’s hard to look at an index for “money” when your policy changes what should be in that index. Longer explication here:http://synthenomics.blogspot.com/2012/05/semi-liquid-money-regime-dependent.html

    Really, on the mechanics of monetary policy, I’m pretty much in agreement with most market monetarism. My only concern with NGDP targeting is its possible propagation of fragilities in other parts of the economy in terms of longer debt chains, etc.

    Of course, from what I see, Major Freedom is just the troll of this blog, so I probably shouldn’t have even responded.

  39. Gravatar of Mike Sax Mike Sax
    13. June 2012 at 14:17

    Major you said Yichuan is no fool. For once you got it right.

  40. Gravatar of Major_Freedom Major_Freedom
    13. June 2012 at 15:00

    Major Freedom, my point about the interest rate doesn’t mean that the current low interest rate environment isn’t contractionary. The current regime is contractionary because NGDP growth is below trend. No, the interest rate in and of itself can’t define the stance of MP, but it in conjunction with other indicators can help define monetary policy.

    NGDP has been growing 4-5% since 2010. I think that’s strong evidence that monetary policy has been loose, which means the current low interest rates are more likely a result of loose money, but again, we cannot say for sure.

    And truly, there’s no way to identify the stance of monetary policy outside of its objective, which market monetarists identify to be NGDP.

    That’s false. We can look at all sorts of objective factors to gauge the extent of how loose monetary policy really is. The fact that some factors would call for more money creation than others, is besides the point.

    Beckworth’s work on safe assets and other work on the problems in the shadow banking center have shown that NGDP growth can have effects on what is even considered to be money. In effect, money is regime dependent; it’s hard to look at an index for “money” when your policy changes what should be in that index.

    Well I recommend you read up on what money actually is, because when you know what money is, you can know what to measure.

    Really, on the mechanics of monetary policy, I’m pretty much in agreement with most market monetarism. My only concern with NGDP targeting is its possible propagation of fragilities in other parts of the economy in terms of longer debt chains, etc.

    As well as accelerating monetary inflation.

    Of course, from what I see, Major Freedom is just the troll of this blog, so I probably shouldn’t have even responded.

    So I guess I can conclude that you don’t know what a troll is. Hint: It’s not someone who disagrees. If you want a troll, you look to see who’s trying to bait into giving emotional responses. For example, the above poster.

    It’s too bad you’ve been fooled by NGDP nonsense.

  41. Gravatar of Major_Freedom Major_Freedom
    13. June 2012 at 15:08

    Mike Sax:

    Major you said Yichuan is no fool. For once you got it right.

    If someone said that about anyone you agree with, you’d say the same thing.

  42. Gravatar of Major_Freedom Major_Freedom
    13. June 2012 at 15:26

    Yichuan:

    I can tell that you too are going to be fun to refute. You have that “aura” of someone who believes he understands the market, but who actually only understands mainstream textbooks.

  43. Gravatar of Mike Sax Mike Sax
    13. June 2012 at 15:36

    Yichuan if Major refuted anything it’d be his first time. And believe me you were right the first time-he is the troll. If there were a vote he’d win in a landslied.

    Major I think I see why you’re so opposed to democracy.

  44. Gravatar of ssumner ssumner
    13. June 2012 at 18:46

    Yichuan, Excellent post. I sent you an email about China.

    MF. I think Yichuan has the good judgment to stay away from any ideas you find appealing.

    Jeff, Just to re-iterate, I agree they know lots of math. But if you read the minutes you see their discussion is based on very primitive models.

    Charlie, There’s a strong bias in many quarters against admitting the obvious, which is that monopoly producers of fiat currency always have lots of influence, the only question is how they choose to use it.

  45. Gravatar of dwb dwb
    13. June 2012 at 18:52

    @ Charlie

    which FOMC minutes is that from?

  46. Gravatar of Major_Freedom Major_Freedom
    13. June 2012 at 19:21

    Mike Sax:

    Yichuan if Major refuted anything it’d be his first time.

    I’ve refuted your claims often enough to make that claim a falsehood.

    And believe me you were right the first time-he is the troll. If there were a vote he’d win in a landslied.

    You too don’t know what a troll is.

    Major I think I see why you’re so opposed to democracy.

    Because I am against initiating force against innocent people?

    ssumner:

    MF. I think Yichuan has the good judgment to stay away from any ideas you find appealing.

    If by “good judgment” you mean thinking like a central planner, I can fully understand what you would believe he should stay away from market based ideas.

  47. Gravatar of A Scott Sumner-Stephen Williamson Dust Up Over Models | Last Men and OverMen A Scott Sumner-Stephen Williamson Dust Up Over Models | Last Men and OverMen
    26. February 2017 at 05:13

    […]       http://www.themoneyillusion.com/?p=14862#comments […]

Leave a Reply