We spend a lot of time thinking about how to get out of a liquidity trap. Maybe we can think more clearly about that question if we instead ask the exact opposite question.
But let's be a little more precise about our question:
How can you get an economy into a liquidity trap in such a way that someone else couldn't get it back out of that same liquidity trap just by reversing whatever it is you were doing? Because it's not really a trap if someone else can get the economy out of it.
I can get a car stuck. That's easy. Any fool can do that. But getting a car stuck so badly that another driver couldn't take over and reverse the car back out is a little bit harder. You have to do something that's irreversible to make it really stuck.
With cars, it's easy to think up lots of irreversibilities. Get it stuck going downhill (it's very hard to get permanently stuck going uphill, because you can nearly always just reverse back down again). Get it stuck resting on a tree stump with all four wheel spinning uselessly off the ground (happened to me once, where the Land Rover's 4WD was finally defeated). And it's extremely easy to get a car permanently stuck if reverse gear doesn't work. What would be the economic equivalents of those sorts of irreversibilities that would make it possible to get an economy stuck and impossible for someone else to get it unstuck?
If monetary policy were loose enough, you would never get into a liquidity trap in the first place. The Zero Lower Bound on nominal interest rates won't be a constraint, for any conceivable shock, if you are targeting 100% inflation. Hitting the ZLB means you were running too tight a monetary policy, given the driving conditions, in the first place.
If one night I snuck in past security, locked the doors behind me, and took over the Bank of Canada, and if I were full of bad intent, I could very easily get the Canadian economy stuck at the ZLB. I would just tighten monetary policy, and announce loudly that I was tightening monetary policy, and force NGDP growth to be so negative that the equilibrium nominal interest rate would need to be negative. No problem.
But what happens when they finally break the locks, escort me out, and Steve Poloz and his old team take over again, and everybody sees that Steve is back in charge of monetary policy?
What could I possibly do, short of physically smashing the printing presses, that would prevent Steve from simply reversing what I had done? Where's the irreversibility?
The fact that I would have destroyed a lot of physical and human capital is a form of irreversibility. But it's not a form of irreversibility that would make it harder or impossible for Steve to get the economy out of the liquidity trap. If anything it would make it easier for Steve, because standard models of investment say that investment demand is a negative function of the existing capital stock, so a war that destroys part of the capital stock causes a rebuilding boom and increases the natural rate of interest when the war ends.
I can only think of one plausible candidate that might work: inflation inertia. If there is inflation inertia, monetary policy still ultimately controls inflation, but the inflation rate has a lot of momentum, so that it's hard to get it to speed up or slow down quickly. If there is sufficient inflation inertia, I might be able to get the Canadian economy stuck in a liquidity trap so badly it would be very hard for anyone else to get it out again.
Suppose I targeted minus 10% inflation, and I eventually hit my target, before they managed to break the locks. If there is inflation inertia, the inflation rate won't immediately go back to 2% when Steve takes over. Just like a car, no matter how powerful the engine, won't accelerate from 0 to 100km/hr in 0 seconds. This means there may be some strictly positive time period during which the ZLB will be a binding constraint on short-term nominal interest rates even after Steve takes over again and everyone knows Steve is back in charge for good. If that is the case, does there exist an equilibrium time-path in which inflation eventually returns to Steve's 2% target? If not, then I have succeeded in getting the Canadian economy irreversibly stuck in a liquidity trap.
The answer isn't obvious. If saving and investment depend only on the short-term real interest rate, and if actual inflation depends only on lagged inflation and current output and short-term expected inflation, then there may not exist an equilibrium time-path in which inflation eventually gets back to Steve's 2% target. Because Steve's forward guidance can influence long-term real interest rates and long-term expected inflation, but these don't matter by assumption. So the short term real interest rate will be above the natural rate for some period, which means output will be below the natural rate for the same period, which means that actual and expected inflation would fall still further during that period, which prolongs the return to the 2% target, and so on. Which means I might be able to get the economy irreversibly stuck.
But is there inflation inertia? The answer to that question isn't obvious either.
There is no inflation inertia in the Calvo Phillips Curve assumed in standard New Keynesian models. If output is at the natural rate, and if expected inflation is at 2%, actual inflation will also be at 2%, regardless of what it was in the past. So if the standard New Keynesian model is true, Steve can reverse what I did and get inflation back to 2% with no lag. But nobody actually believes the Calvo Phillips Curve; we only assume it because it makes it possible to do the math.
Empirically the evidence seems to be mixed, at least to my eyes. During relatively normal times there looks like a lot of inflation inertia. It's just too easy to forecast inflation, especially core inflation, from lagged inflation. But in abnormal times, when there is a clear change in monetary regime that changes expectations, inflation seems to change very quickly. We saw that when the Bank of Canada announced the original inflation targeting agreement. We saw it even more clearly in Sargent's "The ends of four big inflations" (pdf). Dragging me out of the Bank of Canada and putting Steve Poloz back in charge would be a very big and obvious change in monetary regime.
[Update: See Marcus Nunes for another example where apparent inflation inertia suddenly disappeared when the monetary regime changed.]
Dunno. It might be possible to get an economy irreversibly stuck in a liquidity trap. But it's a lot harder than you might think.
Worthwhile Canadian Initiative 
Are we talking about “a situation in which conventional monetary policy — open-market purchases of short-term government debt — has lost effectiveness” (Krugman’s definition of a liquidity trap) or something more lethal?
It depends HOW you create money. If you only create money by expanding private debt – maybe it is hard to get people to take on extra debt. A proper helicopter drop should work – but is that fiscal or monetary policy?
Robert Mugabwe knows how to come out of liquidity trap at the speed of sound and go too far in the opposite direction. But apparently the West’s so called “professional” economists can’t work it out.
Kevin: that definition doesn’t work, because it is silent on the question of whether those are temporary or permanent open-market purchases of short-term government debt. Just like a helicopter drop needs to be defined as either temporary or permanent. I’m talking about something more lethal than just “temporary OMOs or helicopters won’t work”.
reason: almost any monetary policy change will change the profits of the central bank, and if the central bank is owned by the government, those profits sooner or later go to the government, which affects the government’s budget constraint, which has fiscal consequences. The only degree of freedom is in the timing of those fiscal consequences.
Ralph: Yep. Remind us again of Mosler’s Dictum(?) for those reading this, just add that when Warren says “fiscal policy” he means what we call “money-financed fiscal policy”, so he’s really (in our language) talking about a permanent increase in the money supply where the government happens to spend all the central bank’s extra profits right now.
Nick,
http://en.wikipedia.org/wiki/Liquidity_Trap
“A liquidity trap is a situation described in Keynesian economics in which injections of cash into the private banking system by a central bank fail to lower interest rates and hence fail to stimulate economic growth.”
A liquidity trap can be said to be a condition where lowering interest rates does not boost (stimulate) credit growth – yes? Ultimately nominal interest rates and nominal economic growth are positively correlated and so there is a trade off between interest rate reductions and hope by the central bank that credit growth rises.
For private credit growth to take hold both the borrower and the lender must be amicable to the terms of the loan agreement. And so a central bank by pushing interest rates too low can destroy the incentive to lend at those rates by the private banking system – Federal Reserve as lender of last resort becomes lender of only resort.
Nick,
What Mosler and I think most MMTers advocate is combining monetary and fiscal policy. I.e. if there’s a recession, then “create $Xbn of new money and spend it (and/or cut taxes).”
I favour that policy. It does have its POLITICAL difficulties, e.g. it can be difficult to reverse (though the cuts in VAT in the UK over the last four years were reversed without difficulty).
Also, given inflation, the real value of the monetary base shrinks at 2%pa (and much more than that in the UK over the last 5 years). So significant amounts of reversal are not normally needed: rather it’s a case of adjusting the rate at which the base in nominal terms increases.
As to “central bank profits”, I don’t think the phrase means much. E.g. the Bank of England created £60bn out of thin air at the height of the crises for the benefit of two banks, RBS and HBOS. Did that add £60bn to the BoE’s “profits”?
Ralph: “Did that add £60bn to the BoE’s “profits”? ”
Yes. Minus the present value of what they lose if and when they destroy that L60bn again, and plus or minus any profits or losses on what they bought with it.
Nick, I’d say conventional monetary policy is always silent on the question of “whether those are temporary or permanent open-market purchases of short-term government debt.” Any promise that future policy will depart significantly from past practice is unconventional.
“when Warren says “fiscal policy” he means what we call “money-financed fiscal policy”
He argues that “bond-financed” fiscal policy isn’t necessarily less stimulative than “money financed” fiscal policy, it might even be more stimulative.
Kevin: maybe, but conventional monetary policy is not usually silent on the long run target, like whether it’s an inflation target, for example. Announcing a permanent 10% increase in the money supply, for example (so that M(t) will always be 10% higher than it would otherwise have been, for all t after today) is equivalent to a 10% rise in a price level path target.
Philippe: yep, if you pay the interest on those bonds by printing even more money, as opposed to raising taxes.
It’s fascinating that almost no one discusses the possibility that the liquidity trap in the islm model is evidence that the model is wrong. Similar to how the ultraviolet catastrophe was an artifact of rayleigh’s law and not physical. We’ve never seen a liquidity trap happen. Output responded to monetary policy announcements during the depression and it responded during the past five years. Time to start saying the liquidity trap result discredits the old Keynesian models.
Nick,
He separates the taxing from spending because he argues taxes don’t logically fund government spending. What he tends to look at is the overall size of the budget deficit.
Jon: very good point. I sort of said something a bit like that in this old post on “Why don’t we observe black holes?”
“Announcing a permanent 10% increase in the money supply…is equivalent to a 10% rise in a price level path target.”
Yes and I don’t know anyone who would describe such an announcement as conventional monetary policy.
“We’ve never seen a liquidity trap happen.”
John Quiggin seems pretty sure he saw one a few years back.
http://johnquiggin.com/2013/08/26/a-note-on-the-ineffectiveness-of-monetary-stimulus/
Ah but now JQ is trolling:
“AFAICT, MMT is the idea that the economy is always in a liquidity trap (hence unlimited scope for fiscal policy) while MM is the idea that the Great Moderation is forever (hence, omnipotence of monetary policy). Right now, MMT is closer to the truth.”
Inflation inertia isn’t necessary to produce an “irreversible” liquidity trap. If prices are sufficiently sticky relative to the other parameters in the model, then you can end up in an expectations-driven multiple equilibrium scenario where one of the equilibria is a permanent liquidity trap. As you mention, the new keynesian Phillips curve depends on expected short-run inflation, not long run inflation targets, so all is required is a huge amount of price stickiness.
Jon,
“It’s fascinating that almost no one discusses the possibility that the liquidity trap in the islm model is evidence that the model is wrong.”
Not wrong, just incomplete. Newton wasn’t wrong about gravity, he just didn’t have the tools to test his theories at very high speeds (special relativity) or in accelerating frames of reference (general relativity).
Matthew: I’m not sure you are right about that. To my mind, “price stickiness” refers to the level of prices. The price level cannot adjust quickly to the new equilibrium following a change in monetary policy. The Calvo model has price stickiness in that sense. But the price level isn’t relevant for the ZLB. It’s the rate of inflation that is relevant for the ZLB. And “inflation inertia” (inflation stickiness) means the inflation rate cannot adjust quickly.
Maybe I’m misunderstanding you?
Kevin, The period Quiggin discusses Australia was not even close to the zero bound. In addition, he describes a cut in interest rates from 17% to 5% in a period where NGDP growth fell from double digits to one percent as an “expansionary monetary policy” And no, I’m not joking. By the logic the Volcker disinflation (similar interest rate and NGDP data) was an easy money policy!!
http://www.themoneyillusion.com/?p=23215
How would he describe monetary policy during the German hyperinflation? Ultra-tight? My Keynesian commenters kept insisting I was attacking a caricature of Keynesian economics. But it seems the interest rate fallacy is alive and well.
If the economy is in a depression and the government is running a huge budget deficit, would you describe that as ‘tight’ fiscal policy? Or ‘loose’ fiscal policy?
Nick,
“If output is at the natural rate, and if expected inflation is at 2%, actual inflation will also be at 2%, regardless of what it was in the past. So if the standard New Keynesian model is true, Steve can reverse what I did and get inflation back to 2% with no lag.”
I think you are saying that since inflation isn’t sticky, it is possible to change it instantly merely by changing the target. I think that’s true, but it is not something external to the NK model. The target is explicitly embedded in the policy rule and therefore in the NK Philips curve. So it directly controls current inflation via rational expectations of future policy. As I read your post I get the feeling that you see the target as something that can be set independently of the conditional path of the policy rate (the policy rule). I don’t see that.
You have driven the economy into a state of high deflation and high real interest rate. It’s not inertia keeping inflation there, but rather it’s the unique dynamic equilibrium consistent with the current setting of i and pi, and policy and microstructure-consistent expectations (ratex). There is nothing intrinsically reversible about this dynamic. This stick is falling over, not because it has inertia, but because the bottom is too far to the right (nominal rate) and the top is too far to the left (deflation).
So there is no guarantee that Governor Poloz can reverse the damage you did, but it’s not impossible that he can. Within the limits implied by the microstructure of the real economy, the price setting dynamic and the credibility of the available set of policy rules, he can grab the stick by the top and move it a certain amount to the right. If the return of Poloz is a surprise, it produces a sudden regime change, meaning a whole new Philips curve (the Philips curve before will reflect some weighted average of your policy and Poloz’s, depending on the perceived probability of breaking the locks). This will cause a jump in the spot inflation rate, which could, if large enough, result in a sudden transition out of the trap.
But I don’t see how any of this depends on inertia or any mechanism outside of the microstructure of the standard NK model.
Nick,
This is completely off topic but I think have some econometric results that may be of interest to you.
A couple of weeks ago I was commenting at UnlearningEconomics:
http://unlearningeconomics.wordpress.com/2013/08/17/market-monetarism-jumps-the-shark/
And the topic of endogenous money and Granger causality came up. I did some googling and apparently the endogenous money people have been blogging about the empirical evidence supposedly supporting endogenous money for the past few months and it has gone to their heads. Most of this research involves Granger causality.
I’ve been toying with Granger causality, specifically a technigue I read about on David Giles blog invented by Toda and Yamamato. And on a hunch I started running Granger causality tests between the monetary base and broad money during ZIRP episodes involving QE. I’ve consistently found that monetary base Granger causes broad money but not the other way around. When I brought this up at UnlearningEconomics this was pooh poohed because of all the “empirical evidence supporting endogenous money”.
Of course, if a central bank is setting interest rate targets one expects loans to Granger cause the monetary base, and broad money, or the broad money multiplier, which is what some of these studies find (almost all of them are published in the Journal of Post Keynesian Economics). But I’ve noticed in ZIRP episodes with monetary base expansion (QE) normally there is corresponding broad money expansion which the endogenous money people say is impossible.
Well I’ve gone a lot further using US data. What I find is that since December 2008 the monetary base Granger causes loans and leases at commercial banks and that the M1, M2 and MZM money multiplier all Granger cause loans and leases (but neither is the other way around). This is exactly the opposite of what Structural Endogeneity predicts.
I’m thinking I should repeat the work on the UK and Japan and possibly the US during the Great Depression. It sounds very interesting to me of course, but who would publish research showing money is not endogenous during QE using endogenous money research techniques?
“I’ve noticed in ZIRP episodes with monetary base expansion (QE) normally there is corresponding broad money expansion which the endogenous money people say is impossible.”
With QE the central bank buys assets from primary dealers, but these in turn buy them from all sorts of non-bank investors. As such QE directly increases the quantity of bank deposits held by non-banks (broad money), as well as the quantity of bank reserves.
This recent paper details who ultimately sells assets to the Fed as part of QE:
“Overall, our results suggest that the Federal Reserve is ultimately interacting with only a handful of investor types. Households (the group that includes hedge funds), broker-dealers, and insurance companies appear to be the largest sellers of Treasury securities when the Federal Reserve buys these securities. Households, investment companies, and to a lesser extent, pension funds, are the largest sellers of MBS when the Federal Reserve buys. With both the Federal Reserve’s Treasury and MBS purchases, our results suggest that households are the largest, ultimate seller…”
“Analyzing Federal Reserve Asset Purchases: From whom does the Fed buy?” (2013)
Click to access 201332pap.pdf
“This is exactly the opposite of what Structural Endogeneity predicts.”
should read
“This is exactly the opposite of what Accomodative Endogeneity predicts.”
Mark, is Granger causality really applicable? I’m no expert, but it seems to me that whatever the underlying process, it’s probably not purely stochastic. After all, Uncle Ben has adjusted the interest rate dial on the machine – so there is a machine of some description.
Mark A. Sadowski,
That sounds fascinating!
Patrick,
“Mark, is Granger causality really applicable? I’m no expert, but it seems to me that whatever the underlying process, it’s probably not purely stochastic. After all, Uncle Ben has adjusted the interest rate dial on the machine – so there is a machine of some description.”
If you back to Kaldor’s 1970-85 papers/books on the subject, and he’s probably legitimately the originator of the whole Post Keynesian strand of endogenous money, one of the key points he was making contra Milton Friedman was that money demand is unstable. If money demand is unstable you don’t necessarily expect money supply to be the driving nominal variable. Granger causality tests merely confirm that the causality is bidirectional and varies according to time and place, so yes.
Most endogenous money people assume that Market Monetarism (MM), and their associates, are arguing money demand is stable just like Friedman did in the 1960s and 70s. But half a century has passed (Milton Friedman is dead, may he RIP) and yet the endogenous money people are still declaring victory in a war that is no longer being fought. MM et al. is as much, if not more, about money demand but this doesn’t seem to get through the thick skulls of most endogenous money enthusiasts.
Accomodative Endogeneity acknowledges that the money creation process is directly run via the central bank’s short run interest rate control knob. But they also believe that the economy is a driving force in money demand, and if the central bank fails to accomodate that demand by adjusting the dial the economy will “blow up” (whatever that may mean). Most importantly they also believe that once short run interest rates hit the zero lower bound there is absolutely nothing central banks can do. QE cannot affect loans and it cannot affect money supply (i.e. deposits). And yet the Granger causality tests show this is plainly false.
P.S. There are other schools of money endogeneity, namely Structural and Liquidity Preference. Structural endogneity is highly nuanced and I don’t feel up to summarizing their views yet, which to me seem to vary considerably depending on person and time of day. Liquidity Preference merely predicts there is bidirectional causality between loans and money supply which isn’t at all controversial to MM et al.
Thanks the reply Mark, but I was just talking about the applicability of the statistical test from a mathematical point of view. Seems to me the underlying system is likely to be a complex dynamic system, and Granger may not be applicable. Just a thought. I’m certainly no expert.
Patrick,
“Thanks the reply Mark, but I was just talking about the applicability of the statistical test from a mathematical point of view. Seems to me the underlying system is likely to be a complex dynamic system, and Granger may not be applicable. Just a thought. I’m certainly no expert.”
Almost all of the more common criticisms of Granger causality testing are overcome by the Toda and Yamamato vector autoregression (VAR) technique, which is apparently considered state of the art.
However, to be clear, failure to reject the null hypothesis that x does not cause y, does not necessarily mean that there is in fact no causality. A lack of sensitivity could result from too small a sample even if true causation occurs. Furthermore, as its name implies, Granger causality is not necessarily true causality. If both x and y are driven by a common third process, then one might still accept the alternative hypothesis of Granger causality.
I’m very disappointed that so little effort is done to look, quantitatively, at how many loans are given out by lenders (banks, or banks plus others).
Whether it is interest rates or QE, the “tightness” of money is actually how easy or not it is to get a loan. Thus, the measure should be … how many loans.
Perhaps also how many first loans to a first time borrower.
Perhaps also first time business borrower (prior car & house borrower).
And my own suggested cure: Tax Loans. To any taxpayer who paid taxes, an ARM loan equal to the last 5 years of income tax paid, at the Fed interest rate, adjusted quarterly.
Almost a helicopter loan drop — but with counter cyclical long term fiscal impact (spend now in financial crisis, pay back in good times reducing any bubble).
Mark: Interesting. I am very wary of econometric causality tests for a rather different reason: If the central bank is using instrument X to target Y then we ought to find that X does not Granger cause Y, even if X does in fact cause Y, unless the central bank is making systematic mistakes. And even if it is making systematic mistakes, the estimated “effect” of X on Y could as easily have a sign opposite the true sign.
My one foray into Granger causality and money is here (pdf). (It got published in JIMF.)
I would fit your results into my perspective as follows: Granger causality tests only reveal true causality when the central bank is screwing up and making purely random moves.
Karsten: Here’s another way of looking at it (I should have said this in the original post):
Does history matter? If history doesn’t matter, then as soon as Steve Poloz takes over from Nick, it is exactly the same from then on as if Nick had never grabbed control in the first place.
Now history might matter because Nick leaves a lower capital stock, but that won’t prevent Steve escaping the ZLB.
History might also matter if it leaves a higher variance of relative prices across firms, given Calvo pricing and Nick’s stupid monetary policy. But I can’t see how that makes it harder for Steve to escape.
The only thing I can see is if we ditch calvo pricing so the history of inflation matters.
Nick,
“I am very wary of econometric causality tests for a rather different reason: If the central bank is using instrument X to target Y then we ought to find that X does not Granger cause Y, even if X does in fact cause Y, unless the central bank is making systematic mistakes. And even if it is making systematic mistakes, the estimated “effect” of X on Y could as easily have a sign opposite the true sign.”
I’m not sure this is a reason to be wary, it is however a reason to be aware. Unfortunately awareness of how the central bank’s targets and instruments affects causality tests seems to be in very short supply among many of the endogenous money enthusiasts.
P.S. I love your paper.
“Arrow & Fisher (1974) defined an irreversible action as one that is infinitely costly to
reverse” [http://ecoservices.asu.edu/pdf/Perrings%20and%20Brock,%20ARRE%20(2009).pdf]
It is possible to imagine, in theory, reversing actions that would be totally impossible in practice. The costs of reversing an economic action at the macro level may be constrained by many things. For example, if a decision is taken to convert various stored up problems in a financial system into a general inflation or an increase in the stock of money then reversing those actions are constrained, possibly, by things like the zero lower bound and the political possibilities of negative real rates.
The paper above identifies three types of irreversibility:
– “the technical irreversibility of investment decisions lies in the forgone future opportunities
– “irreversibility in the context of the stability properties of different states. Transition
to an absorbing state is irreversible. Transition to a persistent state may be slowly
reversible. More generally, the degree to which transition to some state is irreversible is
implicitly measured by the resilience of the system in that state.”
– “the phenomena
recognized in economics as “lock-in” or “lock-out” are special cases of a more general
property of complex dynamical systems—that their future is entrained by their past.
Feedback effects serve to entrench or exclude some technologies or social processes, at
least for a time.”
Which one of the above are you talking about in this post? I guess its the second one?
scepticus: I think it’s the second one too.
“During relatively normal times there looks like a lot of inflation inertia. It’s just too easy to forecast inflation, especially core inflation, from lagged inflation. But in abnormal times, when there is a clear change in monetary regime that changes expectations, inflation seems to change very quickly.”
Having clarified that its the second regime you are referring to, the above passage seems to refer to ‘critical slowing down’. Critical slowing down (the time it takes for a system to return to its stable configuration following a perturbation which it resists, with the transition-vulnerable system exhibiting longer time to return to the previous trend).
I don’t know much about Granger Causality but I guess it must relate to autocorrelation (where autocorrelation increases with critical slowing down to the transition point).
Anyway, a system with high inflation inertia is one that is prone to return to trend (the previous, existing equilibria) and is thus not exhiniting critical slowing down and is not in danger of an irrevesible transition. It doesn’t need much nudge from monetary policy to do so. If the evidence shows that we have a period of little inflation inertia, that means the transition point is closer, as per the science of phase transitions, complex systems etc.
That suggests that when inflation inertia is high, reversibility is more likely, not less. When inflation inertia is very low, any significant policy action would in theory run more risk of causing a transition to some new equilibria from which it may be impossible to return.
That said, if you look at the 19th century the gold standard then was a stable attractor with considerable resilience but inflation and short term rates were much more volatile than they have been in previous decades. I guess the usefulness of inflation volatility as an indicator of possible reversibility depends very much on which equilibria/regime one happens to reside – it would not have been much use in the 19th century.
On Marcus’s Brazil example: This American Life had a brilliant “The Invention of Money” episode that described it, including interviews with the people who engineered it.
I condense the TAL episode takeaway at the end of this post:
http://www.asymptosis.com/medium-of-account-vs-unit-of-account-brazil-anyone.html
Wherein I also think hard about units/media of account and exchange, among other things coming to a very similar conclusion as JP Koning:
JP: “I think that the current medium of account is CPI units.”
http://jpkoning.blogspot.com/2012/11/discussions-of-medium-of-account-could.html
Moi: “the medium of account is always value”
Ooh yuck that word.
I was kind of astounded when I wrote it to find that monetarists thinkers (bloggers at least) have barely written about the Brazil episode, often aren’t even aware of it — especially astounded given the rather gobsmacking chart of Brazilian inflation that Marcus posts.
Nick, I know you don’t appreciate my work but there is another way to get stuck in the zero lower bound, and the Fed is powerless to get out. Just lower the labor share of income.
Here is a link to an old post with an equation that has changed since, but the idea is there.
http://effectivedemand.typepad.com/ed/2013/05/the-autopsy-of-the-fed-funds-rate.html
Edward: if you assume that workers have a higher mpc than capitalists, then (under certain other assumptions, because it depends on how that would affect investment demand too) a reduction in the labour share of income might lower the natural rate of interest. But that does not mean that the central bank is powerless to take offsetting action to prevent hitting the ZLB. For example, it could simply raise the inflation target.
See that bit in my post where I said: “If monetary policy were loose enough, you would never get into a liquidity trap in the first place. The Zero Lower Bound on nominal interest rates won’t be a constraint, for any conceivable shock, if you are targeting 100% inflation. Hitting the ZLB means you were running too tight a monetary policy, given the driving conditions, in the first place.”?
Did you read that bit? Did you understand it?
Look I know you are really really keen for everyone to read your work. (We are all a bit like that; it’s the bloggers’ sin). But when you jump into all other conversations, just saying “read my stuff!” and adding a link to your blog posts, with all appearances of being deaf to the conversation going on around you, and not having made any attempt to read others’ stuff or understand it, it does get a bit irritating. Because it’s not just here you are doing that. And it makes people even less inclined to read your stuff. Because there’s lots of stuff out there competing for our attention, and someone who has read and understood others’ stuff, and who still thinks he might have something useful to say, is more likely to be right than someone who hasn’t.
Steve: “I was kind of astounded when I wrote it to find that monetarists thinkers (bloggers at least) have barely written about the Brazil episode, often aren’t even aware of it — especially astounded given the rather gobsmacking chart of Brazilian inflation that Marcus posts.”
I think that’s a fair point. And it’s good that Marcus raised that Brazilian point too.
But actually, I think that bloggers, and especially MM bloggers, are much less guilty than most in one very important respect. Me and Scott Sumner actually argue about which is most important: MOE or MOA functions of money? That whole distinction is one that gets very little attention. For example, when New Keynesian macroeconomists talk about “monetary policy” are they talking about MOE or MOA policy?
(I wrote a draft post on this topic a few days back, but it’s not coming out right yet.)
I don’t think you got Mosler quite right Nick unless I’m misunerstanding you. I think for Mosler there is no meaning for central bank’s profit other than for accounting clarity. He consolidetes central bank treasury balance sheets and calls It all government. it’s government money machine to him (the last expression is mine). So he sees central bank higher profit functionally the same as higher net taxes. Your thinking is totally out of paradigm for any MMTer, you think like a monetarist.
You get higher profits for central bank, the more private sector net financial assets are destroyed and government has no budget constraint.
It is very hard to get a car truly and permanently stuck. You just need the right tools to get it un-stuck. And permanent, only in the sense that you still want your car to function. Really there is not a trap deep enough, that someone couldn’t cut your car into pieces and get the car out.
But there are lots of ways to get a car stuck that will require some effort and maybe big tools.
So, to get the economy stuck…we can’t talk about permanently stuck… but we can talk about stubbornly unresponsive…
Merely tightening money doesn’t get the economy stuck. You raise short term rates, and they can be lowered. You increase mandatory bank reserves, and that policy can be reversed. You need money supply to crash while every standard tool of money manipulation is set to easy.
First you need to create the environment when people “over-leverage”? But what is over-leverage. For any degree of risk there is a maximum degree of leverage beyond which increasing leverage does not increase ones expected return. To make people over-leverage you must lower the perception of risk. If volatility stays low enough long enough, it appear rational to lever your book 40x. Then you gradually ease the conventional levels of money supply to make it cheap to lever 40x. Then you need a shock, to make people realize how insane their positions are. The rest will take care of itself. A few people will be wiped out by the shock. Their creditors take a hit, and they call in the loans to others that were damaged but not necessarily dead. This drying up of liquidity pushes the barely alive over the cliff. etc.
We have wiped away the value of financial assets. There is really very little the central bank can do restore value to these assets.
But the funny thing about this, is that not a single real asset was affected. There are the same factories and workers and service professionals. But, without the money-men directing the cash flow work comes to a halt.
Nick, I see you mention MOA and MOA, but what about UOA?
Shoot! replace one MOA above with MOE.
This might have already been said in the comments above(which I just can’t make myself read) but… aren’t you confusing a shift of a particular curve with a movement along a particular curve? Tight monetary policy doesn’t get you into a liquidity trap. An exogenous (real) demand side shock (a shift of the IS curve) does. Of course, successful tight monetary policy which results in near zero inflation means that it takes a smaller exogenous shock (a shift) to get into the trap, but that’s a different thing. And it’s why you can’t just “reverse it”, monetary policy wise, except to the extent that just maybe if you “commit to irresponsible monetary policy” you can substitute higher inflationary expectations for bad animal spirits.
I don’t know about this car analogy. But let’s go with it. It’s sort of like if, for whatever reason, you like to keep your fuel gauge near zero, with the little red light on (maybe because you think it makes you drive in a more efficient, fuel conserving, manner or something). Then an exogenous shocks happens: unexpectedly you find yourself on a long stretch of a road where there’s no gas stations on any exit. And then you’re out of marshmallows. At that point you can’t say “from now I’ll drive with more spare gas in my tank” and have the car magically start again. That’s the irreversibility, the trap. You need to somehow undue to the exogenous shock, get more gas in the tank, then worry about the other stuff.
Like I said, I’m not sure if this car analogy is appropriate but at least it should be gotten right. Fortunately, liquidity trap recessions are not quite like running out of fuel because you were trying too hard to conserve your gas. There’s a spare can of gas in your trunk called fiscal policy (maybe). It’s low grade, diluted with alcohol, environmentally unsound, and half evaporated anyway, but at least it will get you to the next gas station where you can fill up on the real stuff.
(There is some pertinent points and loose ends here though. You’d expect that countries with lower inflation targets fall into the trap more often but is that the case empirically? Are negative supply side shocks really expansionary in a liquidity trap? Do the ZLB models imply a deflationary spiral? Is there such a thing as a NADRU? But that’s beyond the scope of this particular comment and post)
notsneaky: what caused the accident: the sharp bend in the road; or my driving too fast for the bendy road? And if the car is undamaged (if there’s no irreversibility) we can just switch drivers to a slower driver and get out of the ditch, and stay out of it.
Nick: If the ditch is deep enough – a trap – you gonna need a tow truck – fiscal policy – to get you out, regardless of how careful the new driver is. The problem with this version of the analogy is that this would apply to all recessions, not just ZLB ones, unless somehow the depth of the ditch and your prior speed are related.
Aside from being late, this answer could be argued to be semantic. But I think it invokes a useful way to talk about liquidity traps. You ask whether inflation inertia or whether ‘history matters’ might be the only tree stumps we can come up with. Yet these can be thought of as minor components, with many possible friends, to the real culprit, which sadly has to be called credibility.
Chuck Norris can kick Nick Rowe out of the CB. He can exile Nick to Pluto and wipe all citizens’ memories of the incident, or even of the recent inflation experience given the right future version of Google Glass. Now history doesn’t matter, at least in the proximate sense. No more inertia. Yet it may be a much taller order for Chuck to guarantee that a sunspot tight money acolyte doesn’t storm the CB 5, 10 or 20 years hence. Instead of some strange hyper-stylized story about ST rates causing all savings and investment decisions, you keep LT rates in charge as they surely ought to be but simply lose the duration of the CB reaction function. History (in the narrow sense, i.e. a recent regime of tight money) and intense inertia are just two ways to perhaps get stuck in a credibility trap, but require no regime change offset. There are others, including some that have “regime-change” defense built it: Maybe the winds of political change as a fallout from a tight-money amplified recession start to shift toward a party or even future leader that is likely to make a huge mistake in appointing a future CB head, or directly change the CB mandate and prerogative for the (tight-money) worse, years down the road. Or, maybe people are wiped clean of both the Rowe regime and the deflation it produced, but do remember one thing they learned during this time: previously unknown public opinion has been uncovered: people and pundits fear and blame politicians for excessive inflation, bubbles, and pretty much anything unorthodox to such a degree that their worries remains stalwart even when a nominal sledgehammer is pounding the economy. This is the reaction function beneath the reaction function.
Is it unrealistic to worry about the “next” regime change, when some (most??) don’t even seem to think that expectations about contingent future policy matter much in the first place? It is definitely hard. Harder than a more hydraulic story like an “inertia” that overpowers a very long duration path of conditional expectations about the future supply of money relative to demand. But unlike inertia, it could be much “more” irreversible if it requires institutional changes that are either impossible, unlikely or too costly to cure.
Mark, I don’t know all the difference between these two (aside from your descriptions here):
“Structural Endogeneity”
“Accomodative Endogeneity”
But I’ve never heard anyone (including what you might call “endogenous people”) dispute what Philippe referred to here:
http://worthwhile.typepad.com/worthwhile_canadian_initi/2013/08/how-can-you-get-an-economy-into-a-liquidity-trap.html?cid=6a00d83451688169e2019aff1eee5d970d#comment-6a00d83451688169e2019aff1eee5d970d
i.e. that with QE (ZLB conditions) deposits grow in the non-bank private sector since they were swapped for Tsy debt. I had the impression that everyone agrees that “QE is an asset swap.” Is my perception wrong?