On firing your advisers

[Update: Steve Williamson says in comments the lack of communication started much earlier. If that's right, then I think my theory fails empirically in this case.]

If you are in a position of power and responsibility you need advisers. The main job of your advisers is to stop you saying something stupid in public. You say it to your advisers first, in private. If it's stupid, your advisors should tell you it's stupid. That's their job. If they fail to tell you it's stupid, and you say it in public, and the public tells you it's stupid, and you realise the public is right, you should fire your advisers. They have failed to do their job.

Update: you don't fire your advisers because they disagree with you; you fire your advisers because they didn't disagree with you when they should have disagreed with you.

In August 2010, the President of the Minneapolis Fed said something stupid in public.

In May 2013 he again said something stupid in public.

There may be other examples I don't know about.

That's OK. Even very intelligent people sometimes say stupid things. But it's the job of their advisers to stop them saying stupid things in public. And in these two cases, any halfway competent macroeconomist could have advised him that he was about to say something stupid in public. By saying something like this, and this, like I did, only in private.

Noah Smith thinks that [update: Noah in fact says we can't tell whether] Narayana Kocherlakota fired his advisers because he has seen the saltwater light and his advisers are freshwater. Maybe that's part of it. But it's good to have some advisers who see things differently from you, because they might see things you would miss. Provided they do their job.

I think it's much more mundane than that. I think he fired his advisers because he realised they had failed to do their job.

If you model the central bank as setting a nominal rate of interest, then if the natural rate of interest falls, and the central bank does not change the nominal interest rate in response, then there will be excess supply of labour and goods if nominal wages or prices and their inflation rates are sticky. The inflation rate would need to rise to reduce the real interest rate to restore equilibrium. But the inflation rate would actually fall, if there is excess supply of labour and goods. This is a problem with central banks setting nominal interest rates and not responding to shocks. Wicksell knew about this problem. Any halfway competent macroeconomist knows about this problem. If his advisers had been halfway competent macroeconomists, they would have told him about this problem. If they failed to do that, they should have been fired.

This is like a bus driver saying you need to turn the steering wheel clockwise to turn left, and his advisers not pointing out the problem.

In May 2013, he wrote: "I thank participants in a FRB-Minneapolis bag lunch for comments." Unless those comments included something similar to what I have said above, they were useless.

Comments

[JCE]

In the old post you link to, 'Why "everyone" should be forced to take Intro Economics', yo say that Kocherlakota "never distinguished between the equilibrium thought-experiment and the stability thought-experiment."
What did you mean by that?

[Herbert]

@JCE: It seems clear what Nick meant: The Fisher equation i = r + pi describes an equilibrium thought experiment. For instance, a Federal Funds Rate of 3 percent is consistent with a real interest rate of 1 percent and expected inflation of 2 percent.

When you diminish the FFR to 0,25 percent the Fisher equation suggests that an real rate of 1 percent and an expected inflation of -0,75 percent would be consistent. But there is no stable adjustment mechanism that would bring down expected inflation this way. Quite on the contrary, expected inflation would rise upon the reduction in the FFR, and sticking to that policy would eventually result in hyperinflation.

[Nick Rowe]

JCE: an old distinction, that Patinkin made. Here's his example:

Equilibrium experiment: if you double the money supply, all prices will double in the new equilibrium.

Stability experiment: if you double the money supply, but imagine all prices all prices stay the same, there will be an excess demand for goods, so prices will rise towards the new equilibrium.

[Nick Rowe]

Yep. Herbert gives another example of the distinction, and it's the example that directly applies to this case.

[Jon]

Thanks nick. This is the framing NK deserved. Other reports on this topic come off as hatch-jobs. NK is on the bad list: Minneapolis, called for tighter policy, seems right-wing. let's smear him every turn because pk made fun of him once sort of behavior has been seeping out (incl from Noah whose substance seems different but mood afifiliation is clear).

[PeterN]

I'd fire somebody who wrote this

"Consider first the areas of agreement. In terms of methodology we all agree that
in order do serious policy analysis, we need a structural model with primitive interpretable
shocks which are invariant to the class of policy interventions being considered.
Macroeconomic models are also similar. In practice, most macroeconomists now ana-
lyze policy using dynamic stochastic general equilibrium (DSGE) models. These models can
be so generally defined that they incorporate all types of frictions, including various ways
of learning, incomplete markets, imperfections in markets, spatial frictions, and so on. The
only practical restriction from these models is that they specify an agreed-upon language by
which we communicate. A standard aphorism is that if you have a coherent story to propose,
then you can do so in a suitably elaborate DSGE model"

[Nick Rowe]

Jon: thanks. But many of the keynesian/monetarist brethren (of whom I count myself) are more pleased with NK now that he's coming around to "our side". But I still think he makes some basic mistakes, but then, that's what advisers are for, because we all have blind spots.

Weirdly, I am gaining more and more respect for NK as time goes by. According to Steve Williamson, NK made the decision to leave academia and join the Fed because he wanted to make a contribution to public policy, and do the right thing. And he has changed his views in the light of experience, which is hard for all of us. And he is obviously much smarter than nearly all the rest of us who criticise him. And he has had to put up with a lot of flak from people like me. I hope he hires some competent advisers, who can cover his weak spots.

[Herbert]

@PeterN: " The only practical restriction from these models is that they specify an agreed-upon language by which we communicate." This is Kehoe, of course.

In my opinion the more severe restriction that DSGE places on research is that one must eliminate all institutional detail in order to keep the model tractable. Banks and central banks are left out anyway, and in most cases money is also neglected.

Doing DSGE means trading important institutions for perfect optimization.

[Nick Rowe]

Peter N. I wouldn't. Even if (especially if) I disagreed with that.

[Ralph Musgrave]

I’d fire all the advisors (that’s probably the majority worldwide) who clearly don’t undstand Keynes’s dictum: “Look after unemployment and the budget looks after itself.”

I.e. if there is excess unemployment, the deficit needs raising. And worrying about that deficit or any consequent increase in the debt is a sign of economic illiteracy: meriting the sack.

[Nick Rowe]

Ralph: I plead guilty to economic illiteracy, by your definition. And if you were in power, I would be a very good advisor for you, precisely because I would let you know when you were getting this wrong.

[Ralph Musgrave]

Hi Nick,

Then explain where I'm wrong.

Yours, sincerely

Your hypothetical employer.

[Herbert]

Yes, Nick, I seem also to be illiterate because I am convinced that budget constraints hold irrespective of the rate of unemployment.

[Nick Rowe]

Ralph: advising you would be a full-time and very frustrating job. How much does it pay?
(But if they ever put you in charge of UK fiscal policy, I would volunteer for free!)

[Min]

"The main job of your advisers is to stop you saying something stupid in public."

Spoken like a Canadian. :)

[Nick Rowe]

Min: spoken like an ex-associate dean, who has worked for four deans. They wisely *tell us* our job is to stop them announcing stupid decisions, just in case we don't get it.

[Min]

Smith: "The researchers at the University of Minnesota and the Minneapolis Fed have largely hung onto the belief that monetary policy can affect inflation, but can’t fight recessions.

"But there is good reason to think that this view is losing credibility at the Fed.

"Narayana Kocherlakota is...an important bellwether of Fed thinking. His views have shifted decisively toward believing that monetary policy can stabilize the economy. What changed his mind? The answer is obvious: the Great Recession, and the failure of large purchases of long-term government bonds and mortgage-backed assets—QE—to create inflation."

Hmmm. We start from the belief that monetary policy can affect inflation, but can't fight recessions. Then we get evidence that QE fails to create inflation. And our conclusion is that monetary policy can stabilize the economy? Tilt!

Sorry, I couldn't resist. ;)

[Min]

Nick Rowe: The deans "wisely *tell us* our job is to stop them announcing stupid decisions, just in case we don't get it."

Bravo for them! :)

[Min]

About saying stupid things in public:

President Obama on CSPAN, 2009:

"SCULLY: You know the numbers, $1.7 trillion debt, a national deficit of $11 trillion. At what point do we run out of money?

"OBAMA: Well, we are out of money now."

President Obama, State of the Union speech, 2010:

"But families across the country are tightening their belts and making tough decisions. The federal government should do the same. (Applause.) So tonight, I'm proposing specific steps to pay for the trillion dollars that it took to rescue the economy last year.

"Starting in 2011, we are prepared to freeze government spending for three years. (Applause.) Spending related to our national security, Medicare, Medicaid, and Social Security will not be affected. But all other discretionary government programs will. Like any cash-strapped family, we will work within a budget to invest in what we need and sacrifice what we don't. And if I have to enforce this discipline by veto, I will. (Applause.) "

One significant reason for continuing high unemployment in the US is the loss of Federal gov't jobs under Obama.

Meanwhile, Obama's advisors at the time are doing just fine.

[Nick Rowe]

Min: but the guy in charge has to *realise* he's said something stupid in public ;-) And who knows, maybe it was good politics.

[Frances Woolley]

"spoken like an ex-associate dean, who has worked for four deans."

Yup.

Course the punishment for working hard, being loyal and yet still speaking truth to power is getting asked to serve on committees ;-)

[David Andolfatto]

Nick,

I think you're way off base here.
Wish you would have checked with the advisors living in your head before crafting this piece.

First, you (like everyone else) are just speculating. We have no idea what really went down at FRBM.

Second, I do not think NK made himself "look silly" in either of those speeches. An academic tosses ideas out there. The first speech relied on the Fisher equation. True, he did not talk too much about non-RE adjustment dynamics, but still, shame on you for your attempts to bully people for not thinking the "correct" way (i.e., your way). Personally, I do not think it far fetched to believe that the Fed's forward guidance is deflationary (I don't know for sure, but if you call me stupid for contemplating the possibility, even though I understand your counterargument, I would not be too impressed with your views on the scientific method). In the most recent speech you label stupid, all I see are a bunch of slides outlining a model. It is perfectly legitimate to lay out the bones of a model and to discuss its properties -- to ask what the mechanisms are and whether they make sense.

If I had to label anything "stupid," I would begin with this blog post.
And that's what I would tell you as a friend and trusted advisor.

David Andolfatto

[Min]

Nick Rowe: "but the guy in charge has to *realise* he's said something stupid in public ;-)"

After the State of the Union speech, maybe his economic advisors should have resigned.

[JW Mason]

If people want an analytic development of Ralph's post, this post on my blog may be helpful. http://slackwire.blogspot.com/2013/10/functional-finance-and-sound-finance.html

The short version is that it is just as consistent to have the interest rate instrument target the debt ratio and the fiscal balance target the level of output, as it is to have the interest rate instrument target the level of output and the fiscal balance target the debt ratio. In this sense, it is true that fiscal policy never need be subject to a budget constraint.

But it's not quite right to say that "the budget looks after itself." The budget balance is being set by the requirements of maintaining price stability/full employment (however defined). And the debt ratio is being looked after by the central bank. Which it is free to do since the budget authority is looking after output.

If you think this through systematically, you will find that Ralph's policy rule, if consistently followed, would bring the economy to the exact same state as the conventional package of Taylor-type rule plus balanced budgets. Any differences between the rules depend on what kinds of policy errors we think are more likely from the budget authority versus from the central bank. So Ralph's position is more reasonable than you think, but also less radical than he thinks. Or so I think.

[Kevin Donoghue]

On the proper use of advisers:

"Mr Churchill, when he became Chancellor, used to accuse us of giving Hawtrey too little scope. I remember his demanding from time to time that the learned man should be released from the dungeon in which we were said to have immured him, have his chains struck off and the straw brushed from his hair and clothes, and be admitted to the light and warmth of an argument in the Treasury boardroom with the greatest living master of argument."

[daniels]

No Nick, it isn't obvious the inflation rate would fall or what's more relevant it isn't obvious that the inflation rate would fall persistently enough to have a significant effect on expected inflation and therefore on the real interest rate. Your analysis seems based on 1 type of zero lower bound equilbrium, and we can argue that it's not necessarily the most intuitive one. See
http://faculty.chicagobooth.edu/john.cochrane/research/papers/zero_bound_2.pdf

I don't know if this equilibrium is learnable or not. Maybe your favourite reference Howitt 1982 could show it's not, but I'm not sure it matters since standard learning dynamics are probably not very relevant in extreme situations anyways, see for example
http://www.homepages.ucl.ac.uk/~uctpmo0/
.
And the issue with the recent financial crisis is that it didn't conform very well to the Keynesian story with a big deflation. Other factors seemed to be at play that are not part of the standard macro 101 toolkit or part of the typical sticky price Keynesian model.

[Nick Rowe]

David: if NK (or anyone) had said: "This is how nearly all macroeconomists and central bankers think about the relation between nominal interest rates and inflation.......but I think they are all totally wrong, and here's why...." and proceeded to give some sort of argument, even if I thought it was wrong, that would not be stupid.

And academics are allowed to say the stupidest things, and we want them to take those risks, because the downside is very small, and maybe, just maybe, they will say something brilliant if they take those risks. That's their job as academics. But NK is no longer speaking as an academic; he's speaking as one of the bus drivers. And people will freak out if they hear the bus driver giving a very different view of the relationship between steering wheel and which way the bus turns, especially if he seems to be totally unaware that it is a very different view from the one all the other bus drivers hold.

And if I were speaking in any official capacity, representing anyone but myself, I would not have written this blog post without running it past my advisers. And this is obviously just my hypothesis. Nobody knows, except NK, and he's not saying. But my hypothesis fits the facts we do know, and is based on rational behaviour on NK's part. People in authority have advisers, and that's what advisers do, and when they don't do it, it's not surprising if they get fired.

But I appreciate and understand what you are saying, and how you have said it, even though I disagree.

Your friend too.

[Jacques René Giguère]

Nick: two small observations
In "A man for all seasons" , Thomas More says to Chatham:"You made a grave mistake. You told the King what he can do instead of what he should do."
Long ago and very young, I was working for a high-level politician. I once asked him: "You have two economic counsellors,me and X. We always disagree.Why keep both?" "When I decide, I can always say that I did it on the recommendation of my adviser."

[Nick Rowe]

daniels: I really like the work John Cochrane is doing on indeterminacy in New Keynesian models (though I wish I understood it better). But John Cochrane: is aware that other people think differently; is speaking as an academic not as a monetary policymaker; understands that he is talking about what a model says and not what he would do as central banker. Personally, I think the issue is the central bank's strategy space (or communication strategy, if you like). And in a different strategy space, loosening monetary policy *would* mean nominal (and even real) interest rates rising. But if we take the strategy space as is, where it's "setting" a nominal interest rate, that's not what "loosening monetary policy" means.

Those in charge of the policy levers don't naively talk about raising nominal interest rates to raise inflation. And people will freak out if they do.

Yep. I'm surprised that inflation hasn't fallen more given all the symptoms of weak demand. Maybe it's the zero floor on nominal wage changes. Maybe inflation targeting made inflation stickier, so the lag is longer than it used to be. Maybe it's something else. But prices did fall in the 1930's, until Roosevelt's change in strategy space and communication strategy pushed them back up again.

[Jim Sentance]

So we don't get advisors because we're allowed/supposed to say stupid things on occasion? Or is it that nobody cares?

[Nick Rowe]

“Rationalizing the behavior of the president I find impossible,” said Mr. Prescott, a professor at Arizona State University and a senior monetary adviser at the Minneapolis Fed."
I don't see the problem rationalising the behaviour of the president.

Jacques Rene: OK, if you worked as an adviser, to a high-level politician, do you think my description of the adviser's job is correct?

Jim: nobody cares! Plus, we wouldn't listen anyway.

[Jacques René Giguère]

Mostly yep. The problem is that the tsar chooses his cossacks but has no idea how to fight a battle on the steppes.Sohe has no idea what a good cossack is. And the cossack want to fight the Tartars because there is glory in that among the other cossacks...
Henry choose Chatham because he knew he would give him the advice he wanted, especially the bad ones.

[Nick Rowe]

David and daniels: OK, let's forget theory, and look at the empirical evidence:

For the last 20 years, the Bank of Canada has said it is targeting 2% inflation. And it says it has been raising/cutting the nominal interest rate whenever it thinks inflation is heading above/below the 2% target. And for the last 20 years inflation has averaged almost exactly 2%.

If the Bank had been wrong about this, and it should have been turning the steering wheel counterclockwise to make the bus turn right, then it would have been an amazing fluke if the bus hadn't veered off course worse and worse long ago. (Unless the Bank has been lying all along, and secretly knew it had to turn the steering wheel the other way.)

And that's just one country.

Is using the nominal rate as an instrument unproblematic? Certainly not. Half my posts are diatribes against using a nominal interest rate as an instrument. But anyone who does assume the nominal interest rate is the instrument, and thinks you need to raise the nominal interest rate to raise inflation, has got a massive problem trying to explain away the experience of hundreds of bus drivers who think it's the other way around, and who normally manage to steer the bus in roughly the direction they want it to go.

[Steve Williamson]

"I think it's much more mundane than that. I think he fired his advisers because he realised they had failed to do their job."

Do you work there Nick? How would you know that? The truth is that Narayana did not even ask these people for their opinions. As Ellen commented in one of the stories, she has not actually had a conversation with Narayana in years. He has cut himself off from a set of top researchers in the department. They would be glad to give him advice, but he doesn't want it. It's simply weird behavior. Nothing deeper than that.

[Nick Rowe]

Mark Thoma's theory is that NK is unable to herd strong-willed cats. Maybe.

Paul Krugman says something a bit similar to me here:

"One might also want to look at some specifics. There was Kocherlakota’s speech in 2010 in which he argued that low interest rates cause deflation — presumably with some input from the research economists; this was a big embarrassment, he probably noticed, and it may have fed his doubts about whether his economists had anything useful to offer."

[Nick Rowe]

Steve: "As Ellen commented in one of the stories, she has not actually had a conversation with Narayana in years."

Did she say how many years? When did NK stop asking his advisers for their opinions? After August 2010, or before?

[Mark A. Sadowski]

Given Minnesota Macro is such a hotbed of Hallucinogenic Business Cycle Theory (HBC) I thought this might be a good time to review the Wikipedia entry on HBC:

“Hallucinogenic Business Cycle Theory (or HBC Theory) is a class of psychedelic macroeconomic models in which business cycle fluctuations to a large extent can be accounted for by imaginary (in contrast to reality based) shocks. (The four primary economic fluctuations are the gold rush, the bubble (deviation from trend), the counterintuitive movement, and mass hysteria (also known as “the panic” in classic terminology).) Unlike other leading theories of the business cycle, it sees recessions and periods of economic growth as an artificial response to illusory changes in the hallucinatory economic environment. That is, the level of national output necessarily minimizes the irrationally expected utility, and government should therefore concentrate on pretending to make short-run policy changes and intervene through random statements of make-believe fiscal or monetary policy designed to actively and whimsically whip the general public into a false sense of security.

According to HBC theory, business cycles are therefore “hallucinogenic” in that they are based on complete fantasy, and are the most inefficient possible operation of the economy, even given its seemingly perpetually unviable nature. It differs in this way from other theories of the business cycle, like Keynesian economics and Monetarism, which see asset bubbles as being untenable, and recessions as having tangible causes, which lead to what are known as “real-world repercussions.”

An important principle underlying HBC Theory is the principle of irrational expectations. Irrational expectations theory defines this kind of expectations as being identical to a wild guess about the future (a preposterous forecast) that systematically ignores, or thoroughly misinterprets, all of the available information. However, even with an unlimited number of additional assumptions, this theory of expectations indetermination still makes the prediction that human behavior will still be completely capricious and herd like. Thus, it is assumed that outcomes that are being forecast differ arbitrarily or unpredictably from the market disequilibrium results. As a result, irrational expectations differ substantially from disequilibrium results. That is, it assumes that people systematically make errors when predicting the future, and deviations from common sense happen consistently. In an economic model, this is typically modeled by assuming that the expected value of a variable is equal to a spontaneous error term representing the role of ignorance and mistakes plus the value of some completely irrelevant piece of information (such as the price of tea in China).”

[PeterN]

Would you hire/fire this guy?

"But it is still true that monetary stimulus on the whole does not reach the parts of the economy that are in trouble. In the United States solvent households get the privilege of refinancing mortgages at never before seen low interest rates, while households with mortgages “underwater” do not get that opportunity even when they are able to keep up their payments. The ineffectiveness of monetary policy in present circumstances has just about nothing to do with the “zero lower bound”to interest rates that so many economists have agonized about. The reason lies rather in the ages-old maxim of bankers: “Never lend money to people who need it."

"It was one of the lessons of an older brand of Keynesian economics that a disequilibrium arising in one part of the economy will disequilibrate also markets where ruling prices are exactly at the levels that would obtain if the economy were in general equilibrium. In particular, if the rate of interest were above its GE level, one result would be unemployment even at the “right” (GE) level of real wages. Note that downward wage flexibility is unlikely to help in this situation. As long as intertemporal prices are wrong, lower wages will not clear the labor market. If wages were to be very flexible, it would make matters worse. Falling wages and prices would disequilibrate balance sheets in Fisherian debt-deflation fashion.

The point applies with multiplied force if intertemporal markets are not just disequilibrated by a market rate higher than the natural rate of interest but are thoroughly disrupted by a financial crisis. Now, if you are willing to believe that the recent financial crisis either increased the market power of labor or made workers in general lazy, please feel free to stick with GE as the way to interpret the world around you. General interdependence of equilibria is a lot easier to analyze than general interdependence of disequilibria! "

"Representative agent models will not admit fallacies of composition. Keynes taught the Paradox of Saving: if households try to save more than the business sector invests, they will not succeed;instead income will fall. Milton Friedman had his own favorite version of the fallacy: if everyone tries to add to their money balances when the money supply is held constant, most will not succeed; instead, incomes will fall. The fallacy of composition for our times might be called the Fallacy of Deleveraging: if everyone tries to deleverage, most will not succeed; instead asset prices and incomes will fall all around. “The representative agent will not be puzzled by paradoxes of saving; he will not suffer involuntary unemployment; and he is not likely to be gripped by financial panic or to get caught in the maelstrom of debt deflation" Models that do not admit fallacies of composition leave us blind to the major sources of instability in the economy."

I just found an ungated collection of 19 papers by this gentleman.

[Nick Rowe]

Mark: I wonder if that's what I sound like sometimes, to the People of the concrete Steppes!?

PeterN: I don't remember writing that. I don't think I would have written the bit about deleveraging (and monetary policy *does* help those who are underwater on their mortgages), but I have definitely written some of those things he says. But you hire advisers more for what they stop you saying!

"Representative agent models will not admit fallacies of composition."

I'm not sure what he's saying there. Not a clear writer.

"“The representative agent will not be puzzled by paradoxes of saving; he will not suffer involuntary unemployment; and he is not likely to be gripped by financial panic..."

That bit is wrong.

But you are off-topic. So stop now.

[PeterN]

I think this quote from Kehoe explains why people complaining about NK models are being, for the most part, substantively ignored.

"Disagreement stems primarily from differences in model building and assessment. One
tradition, which we prefer, is to keep the model very simple, Disagreement stems primarily from differences in model building and assessment. One tradition, which we prefer, is to keep the model very simple, keep the number of parameters small and well-motivated by micro facts, and put up with the reality that such a model neither can nor should fit most aspects of the data. Such a model can still be very useful in clarifying how to think about policy. Typical examples are the general equilibrium models of optimal fiscal policy pioneered by Lucas and Stokey (1983) which make clear general principles, such as the optimality of smoothing distortions over time and across states. When these models are quantitatively implemented, a simple rule of thumb used to discourage the adding of free parameters is that every time a new parameter is added, some new micro evidence to discipline that parameter should be added as well. Typical examples are the general equilibrium models of optimal fiscal policy pioneered by Lucas and Stokey (1983) which make clear general principles, such as the optimality of smoothing distortions over time and across states. When these models are quantitatively implemented, a simple rule of thumb used to discourage the adding of free parameters is that every time a new parameter is added, some new micro evidence to discipline that parameter should be added as well."

One group knows there's a problem but works around it for the sake of sound microfoundations:

"One tradition, which we prefer, is to keep the model very simple, keep the number of parameters
small and well-motivated by micro facts, and put up with the reality that such a model neither
can nor should fit most aspects of the data. Such a model can still be very useful in clarifying
how to think about policy."

and the other uses complicated models which don't have these problems (things like limits to intertemporal smoothing, a high discount rate for permanent income, uncertainty and precautionary saving).

To them the issue is mostly irrelevant.

Does the undergraduate curriculum address this?

Warning Adobe and Microsoft currently have a disagreement about the glyphs for the character strings "fi" and "ff" on my system (like Öt for fit). I apologize for any of these I missed.

[Nick Rowe]

Peter N: "I think this quote from Kehoe explains why people complaining about NK models are being, for the most part, substantively ignored."

I don't. And he's supporting NC, not NK, models. I read what he is saying as an implied criticism of NK models.

"Does the undergraduate curriculum address this?"

depends on the prof. More likely grad.

I rather liked what Kehoe says there. I appreciate his point, though not sure if I agree.

But this is again off-topic. This post is not an excuse to go off on any track you feel like.

[daniels]

Nick, according to the bank of Canada's core policy New Keynesian model a disinflation must eventually come with lower nominal interest rates to be credible in the medium run once learning dynamics have played out(see figure 8,p. 79)
http://www.bankofcanada.ca/wp-content/uploads/2010/01/tr97.pdf
. Now, bear with me while I take seriously for a second their loglinear model approximation and reverse the dynamics to say that an increase in the inflation target must eventually lead to higher nominal interest rates (holding constant other shocks that would would reduce long run real interest rates such as a permanent increase in pessimism or risk aversion, or a permanent worsening of credit constraints due for example to tighter financial regulation).
In the short run, under imperfect central bank credibility (or partially backward expectations) the policy interest rate does indeed have to go in the opposite direction, but eventually it has to reverse course.
Why is this so surprising? If there's a big negative shock then stock prices must fall a lot in the short run to provide an expectation of future increases (empirically there is momentum in stock prices when looking at 1 year horizons, but mean reversion when looking at 5 year horizons). A similar comment could be made on exchange rates where uncovered interest rate parity works much better at e.g 5 year horizons than in the short run I think. In all these cases, 1-2 year dynamics should not be confused with 3-5 or 10 year dynamics.
Otherwise you could end up like Japan where a commitment to quasi permanent zero interest rates leads to deflation, once short run animal spirits and misperceptions have played out.
As for explaining big recessions without deflation, there are many models of that. You can link financial and confidence shocks to reductions in the efficiency of matching in markets when fewer people are searching for goods, higher uncertainty or pessimism reduces both capital investment and hiring (which is like an investment in a world with labour adjustment costs), big shocks to certain sectors like housing or financial services lead to lower output just because of intersectoral adjustment costs, all the previous channels get amplified in an economy with nontrivial input output structure where intermediate goods are a key part of production. And all these channels work in real business cycle models where nominal variables are indeterminate and there's no Phillips curve in sight (it could be there, but it's not a structural feature of the economy). I can't summarize the state of the art modern business cycle theory in a blog comment for you. You (and a lot of Keynesian sticky price analysts who have misspecified their aggregate supply) have some catching up to do.

[Ralph Musgrave]

J.W.Mason,

You say “it is just as consistent to have the interest rate instrument target the debt ratio and the fiscal balance target the level of output, as it is to have the interest rate instrument target the level of output and the fiscal balance target the debt ratio.”

My hunch is that those two options cannot be equally good, and in fact I favour a system where interest rate adjustment plays no role at all (expect perhaps in emergencies). That’s for various reasons for example it strikes me that the interest rate is simply the price of something, and like all prices, it’s best left to market forces. Second, if stimulus is implemented via interest rate cuts, then stimulus is channelled into the economy just via extra investment, and that’s distortionary: it And makes no more sense than channelling via car production and restaurants.

Certainly Milton Friedman and Warren Mosler advocated a system where there is no government debt: government issues just monetary base. In that scenario, the government / central bank machine would not influence interest rates. Plus Keynes said that in a recession, printing money and spending it was a perfectly good alternative to borrowing money and spending it.

Of course I realise that Nick Rowe won’t approve of the latter ideas. If I was his advisor I’d get the sack.

[Nick Rowe]

daniels: I was following you fine up till you said "Why is this so surprising?" Because everything you had written before that wasn't surprising to me. But I couldn't quite get what you were saying after that.

Let me try it this way: suppose the Bank of Canada suddenly announces it has increased the inflation target from 2% to 3%. We know that nominal interest rates will eventually be (roughly) 1% higher than they would have been. But what happens to nominal interest rates immediately? I don't know. But I think it is perfectly possible they will immediately rise (or need to rise to prevent a boom), if the announcement is credible, and people hear it, so it becomes a (excuse the metaphor) "game-changer", like a change in a Schelling focal point, so that people and firms suddenly see the "rules of the game" have changed, and all change how they set prices and value assets. And we do see sudden/immediate changes like that when there is a currency reform. France didn't have a massive recession when they swapped the Old Franc for the New Franc, so the money supply fell to 1% of its former value. Everybody immediately coordinated on the new equilibrium. It was a purely nominal change, where everybody understood what the new names meant. That sounds obvious to us, but we need to remember that all monetary policy changes are really just changes in naming conventions.

Whether nominal interest rates would immediately rise in any particular model will depend on the Phillips Curve in that particular model, and how expectations change in that model.

But there is a big difference between thought-experiments like that, and the Bank of Canada raising nominal interest rates with no announced change in the monetary policy target/regime. The very same physical actions can have very different consequences, depending on what you tell people you are trying to accomplish by doing those actions. If I slap you on the back, am I congratulating you, swatting a bug, or trying to start a fight? When I raise nominal interest rates, am I trying to prevent inflation rising above the 2% target, or am I raising the inflation target to 3% and trying to prevent the boom that would occur if I raised the inflation target and didn't raise nominal interest rates at the same time?

But again, if 99.99% of central bankers interpret "raising nominal interest rates" as "tightening monetary policy to prevent inflation rising above the existing target", any central banker who says it is loosening monetary policy and part of a change in the monetary regime to a higher inflation target has to realise he is speaking a very different language and make very clear he is speaking a different language.

I am much more amenable than most economists to the idea that raising nominal interest rates might *mean* (as opposed to "cause", because what we are talking about here is the *meaning* of actions, because actions are like words) higher inflation. Interpretation matters. But when NK said what he said, he seemed to be blissfully unaware that there is even an issue here, and that everyone else speaks a totally different language. If he had talked about increasing the growth rate of the money supply, he would have been fine. But M-language is a totally different language from i-language. And central bankers (unfortunately) all speak i-language, and have taught us to speak and think in i-language. And when I say M-language and i-language I am not speaking metaphorically. These are languages. The same physical behaviour means different things and has totally different consequences, depending on what language the people in the economy speak and assume you are speaking. Monetary policy is 99% communications strategy.

Yep, financial crises can have supply-side effects as well as demand-side effects. That is another possible explanation for why inflation failed to fall.

[Nick Rowe]

Ralph: "Certainly Milton Friedman and Warren Mosler advocated a system where there is no government debt: government issues just monetary base."

But Milton understands, and I don't think Warren does, that this means (in a world where currency pays 0% interest) deflation equal to the natural rate of interest.

There are 3 instrument: G, t, and M.

Set G so that MU(G) = MU(C) + marginal deadweight costs of distorting taxes.

Set t to equalise the discounted marginal deadweight costs of distorting taxes over time and ensure the intertemporal government budget constraint is just binding (plus intergenerational equity considerations in an OLG model).

Set M to get AD right.

[Nick Rowe]

JW and Ralph: BTW, I'm pretty sure that neither Milton Friedman nor Warren Mosler are communists, but communism is precisely where their proposal would lead!

[JW Mason]

Functional finance of the Lerner-Modler type rejects that rule. Instead you set taxes so that aggregate expenditure equals potential output. With a budget authority following that rule, the natural rate of interest is always the current interest rate, whatever that may be. So no deflation is implied.

[JW Mason]

Sorry, functional finance rejects this:

Set G so that MU(G) = MU(C) + marginal deadweight costs of distorting taxes.

[JW Mason]

Sorry, that's wrong. There is no difference between the assumptions of
functional finance and conventional policy on this point. If you are free to choose G - T to keep the government debt-GDP ratio on its desired path, then you must also be free to instead set G - T to keep aggregate expenditure on its preferred path.

[Steve Williamson]

Nick,

Lack of communication with the people in question started when Narayana took office. First, he hired a research director from outside, and announced at his first meeting with the research department that he was going to run recruiting. This, after no discussion with people in the department. He made it clear early on that he was not interested in policy advice, from Ellen, for example.

[Nick Rowe]

Steve: if that's right, then I think my theory fails empirically. (Unless he already knew from the beginning they wouldn't be able to do the job they needed to do. But even then it's a puzzle why he didn't fire them earlier, if he didn't want their advice.)

[Jon]

Sorry, but when a new leader joins a big organization what Steve describes is common, even if NK was perhaps clumsy about it given the description.

For instance, new CEO came aboard. Senior leadership went on a retreat. CEO identified power rival in the org. He set the agenda at the retreat: discuss how it is hard to work with this other guy. Ostensibly this was to help the other guy. But it was also to establish a power hierarchy and reduce the others guys standing--signaled it's okay to not align with him.

This was smoothly done in my example-- I only realized what happened and why when the CEO explained his coup to me a few years later.

Firing your rival can work, but it can also be counterproductive because it isn't an organic shift in the power structure; it's purely exercising positional power

[Ralph Musgrave]

Nick,

I’m baffled by your argument in the above link to the effect that a Friedman/Mosler regime (where the government / central bank machine issues money, but no debt) would be communist.

You start by saying “suppose we had a very high long run growth rate in the stock of money supplied by the central bank (CB), and a very high inflation rate.” Then you say that in that scenario, the private sector would hold a relatively small stock of money. True: obviously holding money given high inflation is not smart.

You then say that given declining inflation, the private sector would hold a larger stock of money. Agreed. Then you say that as inflation drops to near zero, the size of the central bank expands enormously and approximately equals all the assets in the economy because the private sector would want an astronomical amount of central bank money.

I have three problems with that argument, as follows.

First, there is no need for a CB or indeed the CB / government combined to hold assets that equal its liabilities. As Warren Mosler pointed out, a CB is like an umpire in a tennis match: handing out points that are of value to players and which the umpire produces from nowhere. That may contravene the rules of double entry book-keeping, but book-keeping entries are irrelevant: the important factors are real effects on the real economy, like the fact that excess money printing by the CB / government produces inflation.

A CB is not like a commercial bank, which I agree has to have assets to back its liabilities. For a CB or CB / government combined, the value of their liabilities lies in the fact that those liabilities are the dominant form of money in the country concerned. And that in turn derives from, first, the fact that money has been declared legal tender, and second, the fact that private sector agents must acquire a stock of that money so as to pay taxes – else they go to prison. And that’s a very persuasive reason.

Second, why would the private sector want an almost infinitely large stock of CB money? Households want enough money for the well known precautionary and transaction motives, and that’s about it. Any more money they get will get turned into a new car or a larger house or world cruise.

Third, you take the argument further, and consider what happens when inflation becomes negative: i.e. there is a return (in real rather than nominal terms) to be earned on holding central bank money. I think that is entering the realms of fantasy. It’s generally accepted that a positive 2% rate of inflation is optimum (which I agree with).

[Nick Rowe]

Ralph: "It’s generally accepted that a positive 2% rate of inflation is optimum (which I agree with)."

Milton Friedman's essay (as opposed to Friedman himself) disagreed.

If you haven't read his essay, you might not get my point. Try this:

Inflation is like a negative interest rate on currency (which pays no explicit interest). 3% deflation is like paying 3% interest on currency.

What is the socially optimum rate of interest on currency? People are prepared to hold currency, despite it's paying a lower interest rate than other assets, because it is more liquid. But it costs the government nothing (ignore paper and ink etc.) to provide that liquidity service. Therefore (on the same basis as the optimality of marginal cost pricing), the government, as monopoly supplier of currency, should price that liquidity service at zero. Therefore it is socially optimal for currency to pay the same rate of interest as other assets. But if people always prefer more liquid to less liquid assets, this requires people to hold only currency. And the central bank has to hold all the other assets, and use the interest it earns on those other assets to pay the interest on currency. Zero-profit central banking.

[Nick Rowe]

Assume (for simplicity) a monopolist has horizontal MC and ATC curves, where MC=ATC at all outputs. It is socially optimal for that monopolist to earn zero profits. So we want the central bank to earn zero profits on its business. Assuming MC=ATC=0 for the central bank, it must never print money, unless it pays interest on that money equal to the interest it earns on the assets it buys with that money. Otherwise it will earn positive monopoly profits.

[Min]

Nick Rowe: "But if people always prefer more liquid to less liquid assets, this requires people to hold only currency. And the central bank has to hold all the other assets, and use the interest it earns on those other assets to pay the interest on currency."

People don't always prefer more liquid to less liquid assets, do they? Many prefer land. Others prefer gold. Others prefer huge round stones. :) For others the optimal liquidity is more than 0 but less than 100%.

Nick Rowe: "it is socially optimal for currency to pay the same rate of interest as other assets."

Are you sure about that?

Nick Rowe: "People are prepared to hold currency, despite it's paying a lower interest rate than other assets, because it is more liquid."

If so, why not let them? Are they wrong?

[Nick Rowe]

Min: "People don't always prefer more liquid to less liquid assets, do they?"

For the same rate of return, and other things equal, yes (except for the people who put their credit cards in the freezer). But maybe other things aren't equal.

"Are you sure about that?"

I'm sure about nothing. But the social optimality of marginal cost pricing isn't something I just thought up.

"If so, why not let them?"

But that's the whole point of my argument, that we *should* let them. Which ends us up in the bizarre conclusion that the central bank should hold all the assets in the economy!

[Nick Rowe]

Basically, to implement Milton Friedman's optimal quantity of money, we stay at the ZLB forever, and the central bank does a truly massive QE to buy any asset that people want to sell and keep us at full employment despite the ZLB!

[Majromax]

> Assuming MC=ATC=0 for the central bank, it must never print money, unless it pays interest on that money equal to the interest it earns on the assets it buys with that money. Otherwise it will earn positive monopoly profits.

That's a problem already dealt with, though. The Federal Reserve earns interest on its holdings, but its profits are remitted to the Treasury. To the extent that its "holdings" are (were) largely government debt instruments, that also made distributional sense.

[Majromax]

... of course, doing that along with unlimited satisfaction of demand for money goes right back to communism. Serves me right for first reading the quoted post in the context of the Say's Law thread rather than here.

[Min]

Nick Rowe: "it is socially optimal for currency to pay the same rate of interest as other assets."

Moi: "Are you sure about that?"

Nick Rowe: "I'm sure about nothing. But the social optimality of marginal cost pricing isn't something I just thought up."

Nick Rowe (earlier): "It is socially optimal for that monopolist to earn zero profits. So we want the central bank to earn zero profits on its business."

But the CB, as part of the gummint, is special. It has an indefinite amount of money. Therefore profit and loss are meaningless for it. To speak of the CB earning a profit is to commit a category error. (Yes, I know that people do. Sigh.)

Nick Rowe: "People are prepared to hold currency, despite it's paying a lower interest rate than other assets, because it is more liquid."

Moi: "If so, why not let them?"

Nick Rowe: "But that's the whole point of my argument, that we *should* let them. Which ends us up in the bizarre conclusion that the central bank should hold all the assets in the economy!"

I don't think that I was clear. I meant, Why not let them hold currency that pays a lower interest rate than other assets. It could even pay negative interest, like Gesell's money or store coupons. Use it or lose it.

Anyway, you assume that the CB should pay interest on currency because that would be socially desirable, and end up with a conclusion that seems to be socially undesirable. A nice reductio ad absurdum, eh?

[Ralph Musgrave]

This sentence of Nick’s has caused some interest above: “Therefore it is socially optimal for currency to pay the same rate of interest as other assets.” I don’t agree with that sentence.

I agree that “the government, as monopoly supplier of currency, should price that liquidity service at zero”. But if private sector entities are going to hold assets over and above their stock of money (ignoring personal assets like houses and cars) they’ll quite reasonably want a reward for doing so, and for two reasons. First because of the risk involved, and second because they’ll have forgone consumption in order to purchase those assets.

Offering a reward for forgoing consumption seems perfectly reasonable and “socially optimum” to me.

[Nick Rowe]

Ralph: "Offering a reward for forgoing consumption seems perfectly reasonable and “socially optimum” to me."

Yes, it (usually) is. Suppose that reward should be (say) 3%. Then 3% deflation means that people holding Bank of Canada currency, which pays 0% nominal interest, also earn a 3% real interest reward for postponing consumption.

[Jacques René Giguère]

Whether your adviser is « rightist » or « nerdy » is rarely that important. During the Cold War, lots of research was done by scientist from NATO-WarPact on weird things from 63 Mt bombs to nuclear grenades and various horrible gases. Not counting what we never knew and probably never will (Russians are still mighty scared of microwaves).
Some did it to exterminate the godless-communists-who-pollute-our-precious-body-fluids or the running-dogs-lackeys-of-international-capitalism or to prevent being exterminated by same. Some did it by patriotism for the-land-of-the-free or mother-Russia, irrespective of the political regime. Many did it because it was a well-paying job. Most did it because it was a cool project.
Most of the time, their inner motivation was irrelevant. Most important is the motivation of the decision-maker.

[Jacques René Giguère]

The tsar chooses his Cossacks even though he doesn’t know how to fight on the steppes.
The principal has to choose advisers. But how? If he needs advisers, it’s because he doesn’t know about the subject. How can he evaluate the candidates or the value of their advice? By asking somebody else? How can he choose them? This is an infinitely recursive problem: “Quis custodiet ipsos custodies?”
So Barack Obama chooses Tim Geithner whose own objective is to save the U.S. economy by saving Wall Street. Or maybe just saving Wall Street. Or just Wall Street bankers. How am I supposed to know?
Your primo probably know nothing. Worse, sometimes he thinks he knows something. In economics, almost anyone from the business world will be a catastrophe. The best compliments we could bestow on a minister was: ”He could be deputy minister.” Even better: “He could be his own deputy minister.” Civil servants hates them because a) they can outwit their advisers and b) their example let other ministers think they can outwit their staff...
The primo’s knowledge is limited, And he may say stupid things “We’re running out of money”. Or “We need to tighten our belts.” It’s stupid in the context of the economic adviser. But are they stupid in the context in which the primo works? The primo work in the world of politics.
I tell the primo what it should do. He knows what he can do. Yes maybe he should teach the public. But there are limits to what the public wants to hear.
Walter Bagehot wrote in “The character of Sir Robert Peel”
http://oll.libertyfund.org/?option=com_staticxt&staticfile=show.php%3Ftitle=2260&chapter=212967&layout=html&Itemid=27
that Peel never was in favor of an idea before the English bourgeois found it obvious (I condense the chapters…).
What Obama said was not stupid but he has to take into account the limits of what the people are ready to hear and understand.
Last Thursday, Québec Minister of Finances Nicolas Marceau published his Economic update. An economically well-versed journalist asked him why his target for budget equilibrium was so fast. “Because the public expect it.” About as stupid as an economist can get. He knew it was stupid But you could see the sadness in his face. It was the only political answer he could give…
This the world the adviser work in.

[The Hat of the Three-Toed Man-Baby]

Good lord Ralph Musgrave is an idiot.