A guest post by Mark Sadowski
Simon Wren-Lewis is at it again:
While the reasons for the Great Recession may still be controversial, the major factor behind the second Eurozone recession is not: contractionary fiscal policy, in the core as well as the periphery. So this is something we really do know.
It is only because of the continuous issuance of preposterous statements like this by people who should know better, that the causes of the Great Recession are even controversial. Not only is this something we do not know, it is something that we know is not true.
The second Euro Area recession was clearly monetary in origin. The ECB raised the MRO rate from 1.0% to 1.25% in April 2011 and then to 1.5% in July. The six quarter second recession started the following quarter. The Euro Area was never near the zero lower bound in interest rates and the ECB chose to raise the policy rate in the face of projections that core inflation would remain below target and the output gap would remain high for at least the next two years.
The US makes a useful comparison. The Fed kept the fed funds rate near zero throughout 2010-2013 and elected to do a second round of QE in late 2010 to mid-2011 and to start a third round in late 2012. The ECB has yet to initiate even a single round of QE.
According to the April 2014 IMF Fiscal Monitor the US increased its cyclically adjusted primary balance (CAPB) by 4.7% of potential GDP between 2010 and 2013 (bottom half Table 1.1 page 11):
The five core Euro Area nations that Simon Wren-Lewis considers (France, Netherlands, Belgium, Austria and Finland) increased their CAPB by a weighted average (according to 2010 nominal GDP) of 2.6% of potential GDP from 2010 to 2013. The GIIPS increased their CAPB by a weighted average of 5.0% of potential GDP from 2010 to 2013.
Between 2010 and 2013 nominal GDP (NGDP) increased by 12.3% in the US, by 6.2% in the core Euro Area nations and decreased by 1.8% in the GIIPS. Between 2010 and 2013 real GDP (RGDP) increased by 6.6% in the US, by 1.4% in the core Euro Area nations and decreased by 4.1% in the GIIPS.
It would have made absolutely no sense for any of the Euro Area nations to be doing fiscal stimulus when the ECB was nowhere near the zero lower bound and in fact raising the policy interest rates. Moreover the US economy has easily outperformed the economies of all of these countries in both nominal and real terms despite being at the zero lower bound in interest rates, and doing roughly the same amount of fiscal austerity as the GIIPS.
The one thing that the Great Recession, in tandem with the Great Depression, has made absolutely clear is that the liquidity trap is a myth promoted by those with a agenda no matter how much empirical evidence there is to the contrary.
Historinhas 
Pingback: The fiscalists vs the monetarists: lacking a common language | Freethinking Economist
The interest rate is not so much the figure of merit as, say, a Taylor rule — i.e. a negative Taylor rule indicates a liquidity trap. Interest rates at the zero lower bound with no inflation imply a liquidity trap, but the converse is not necessarily true … otherwise the Fed could raise rates to 5% and end the liquidity trap.
Additionally, the monetary offset (effective monetary policy) picture and the liquidity trap (ineffective monetary policy) picture are strongly model dependent.
http://informationtransfereconomics.blogspot.com/2014/05/models-matter.html
The two sides of this argument appear to be talking past each other, neither entirely understanding the other side’s argument or the implicit assumptions their own side is making (e.g. Krugman describes the market monetarist model incorrectly and e.g. Sumner describes the liquidity trap argument incorrectly).
http://informationtransfereconomics.blogspot.com/2014/05/monetary-offset-what-are-assumptions.html
Excellent post.
The one thing that the Great Recession, in tandem with the Great Depression, has made absolutely clear is that the liquidity trap is a myth promoted by those with a agenda no matter how much empirical evidence there is to the contrary.
Hear, hear.
Even QE qua QE is a weak policy as it seems to mainly serve as a signal of the path of future CB policy. I remember Tyler Cowen being very surprised by a large market movement in response to a small but unexpected QE announcement. That response makes little sense as a reaction to QE qua QE, but makes perfect sense as a response to QE as affecting market expectations for the Fed’s future behavior. Unfortunately “surprise” is difficult to metric, but this does suggest NGDPLT or other ways of changing expectations would be more salubrious.
But I suppose QE is what we have, certainly the ECB has been bewilderingly tight either way. I haven’t read any ECB minutes, are they worried about inflation right now?
“otherwise the Fed could raise rates to 5% and end the liquidity trap.”
The Fed could accomplish this very easily simply by announcing a long-term 7% inflation target. That’s why there’s no such thing as a liquidity trap, it only exists if you assume the Fed cannot do things the Fed clearly can do.
“The Fed could accomplish this very easily simply by announcing a long-term 7% inflation target.”
Yes, that is the assumption in the monetarist model. The “expectations trap” model is that with rational expectations current inflation equals the average expected future inflation E[i] = i* + ε over the few years (i.e. inflation will average, say, i* = 2%) so that if the Fed announces a 7% inflation target, that means 7% for one year then 1% for a couple of years after that until average inflation is 2% over the period (i.e. the expected value). The monetarist model implies that the Fed announcement changes expected inflation to E[i] = 7% + ε indefinitely, and assumes no one thinks the Fed will “take the punch bowl away” (this is Sumner’s explanation for why there hasn’t been any inflation from QE — the Fed is expected to undo it; that goes for QE in Japan as well).
This is an example of both sides talking past each other. They have different models, with equal amounts of empirical support, i.e. zero. Macro data is uninformative, so we can’t distinguish these models. I personally don’t believe expectations explain anything. That is, I think it is an ad hoc model that allows one to assume the answer one wants — see the paragraph above! I tend to believe incorrect expectations lead to lower growth, regardless of whether they are incorrect optimism or incorrect pessimism:
http://informationtransfereconomics.blogspot.com/2014/05/the-effect-of-expectations-in-economics.html
Yes, that is the assumption in the monetarist model.
It’s not an assumption, it’s inherent in the definition of a monetary sovereign.
They have different models, with equal amounts of empirical support
There is 100% empirical support for the notion CBs can inflate at will. No CB has ever tried to inflate and failed. If you know of a case where that was not true, please do share!
I personally don’t believe expectations explain anything.
You don’t believe in futures markets? I don’t even know what to say to that.
the Fed is expected to undo it;
I think you’re missing the point of forward guidance — the Fed is not just “expected” to undo QE, the Fed is PROMISING that QE will not be inflationary in the long run. It’s a bit of a schizophrenic policy, as noted above.
If the Fed decides on, say, a 10% inflation target, it only has to do enough to make the markets believe it, because everyone betting against the Fed will lose, and investors don’t like to lose. I think it was Lars who christened this the “Chuck Norris effect.” If Chuck wants to clear a room, he doesn’t have to beat up everyone in it, he only has to threaten to beat up anyone who doesn’t leave, and do enough to make the promise credible. Since everyone knows Chuck is invincible (much like a monetary sovereign trying to inflate), they leave the room with only a few punches thrown.
@talldave2
A definition is an assumption.
And I do believe in futures markets in the sense of performing the function of allocating resources, but saying the price in a futures market is what it is because it is expected to be what it is plus some random error doesn’t explain the price given by a futures market. It just moves one’s ignorance from the question “why is the price what it is?” to the question “why is the price expected to be what it is?”
Pingback: Liberalism unbound: Free lunch and dinner--all you can eat!, by Scott Sumner - Citizens News
The definition of definition is “a statement of the exact meaning of a word, especially in a dictionary.”
The definition of assumption is “a thing that is accepted as true or as certain to happen, without proof.”
If a CB cannot inflate, it is not a monetary sovereign, it is something else — by definition. That’s not an assumption, that’s assigning meaning to a word and applying some basic logic.
Anyways, word games are silly. Do you really want to argue that sometimes CBs can’t inflate? Let me know when you find an example…
If you believe in futures markets, then you must either believe the Fed can affect markets by setting expectations, or you must explain why the Fed cannot set expectations — which takes us back to the notion they cannot inflate at will.
I think one can certainly argue that the markets may not believe the Fed will do what the Fed claims it will do, or that there are political considerations that will affect Fed policy decisions, but that’s an entirely different assertion than one that the Fed cannot do what the Fed claims it will do.
The liquidity trap model rests on the reluctance of CBs to inflate, and little else. It’s not a firm foundation.
@talldave2 These are not word games — you need to understand the assumptions being made in the monetarist model. You define a monetary sovereign as an entity that has a CB that can inflate. You assume an axiom that a monetary sovereign can always inflate. When you ask: “Do you really want to argue that sometimes CBs can’t inflate?” you are demonstrating that “a monetary sovereign can always inflate” is an axiom: a proposition so self-evidently true that it needs no proof. In terms of logic:
P = inflation is x%
Q = CB is targeting x% inflation
Axiom P→Q (assume a CB can always get x% inflation)
Observe P (inflation is x%)
Therefore Q (CB is targeting x% inflation)
It’s totally logically consistent (and even consistent with empirical data), but it doesn’t explain anything. It takes the condition of the world as it is and says that the world is that way because it has to be that way.
The BOJ, the Fed and the ECB represent three examples of CB’s that can’t inflate at will — they are all undershooting their inflation targets. However if you assume P→Q above, then by assumption these CB’s are not undershooting.
Here is a model that says a CB can’t always get x% inflation that is broadly consistent with empirical data in the US and Japan:
http://informationtransfereconomics.blogspot.com/2014/03/the-diminishing-effect-of-monetary.html
… it is evidence that P→Q is false sometimes. A different model — the liquidity trap — does not assume P→Q. Maybe these models are wrong, but they are stark evidence that P→Q is a model assumption.
You define a monetary sovereign as an entity that has a CB that can inflate.
I don’t define monetary sovereign that way, everyone does.If it’s an assumption, it’s an absolutely bedrock assumption, on par with physical constants like the atomic weight of boron or the Planck constant.
But like I said, I don’t care for word games. If you don’t think a CB can inflate, then try to defend your argument.
The BOJ, the Fed and the ECB represent three examples of CB’s that can’t inflate at will
None of them are trying to inflate! Look at their inflation targets.
It’s trivial to demonstrate that “can’t inflate” is totally false. Let’s say the Fed decides tomorrow they want a 7% inflation target for 2015. They’ve seen your model and they expect the base money increase necessary to be large . In response to concerns about Cantillon effects, they annnounce they will implement the following: starting on Jan 1 2015, on the first of the month they will send every American citizen cash. They will then monitor CPI and other price level indicators, and if they do not get the response they are looking for they will double the amount the next month, until they reach their target.
You and I might disagree on how large the cash payouts would have to get, but I think we must agree that at some point they would get their inflation target.
Here is a model that says a CB can’t always get x% inflation that is broadly consistent with empirical data in the US and Japan:
What you’re measuring there is the effect of promises not to inflate. It’s circular reasoning: the CB cannot inflate because it has decided not to inflate. That’s the problem with empirically-derived models of people’s decisions: just because Joe drove to work every morning for 30 years doesn’t mean he can’t decide to do something else tomorrow.
they are all undershooting their inflation targets.
If you read the Fed minutes you’ll know why, they prefer to miss low. Heck, in fall 2008 they were still worrying about inflation. I don’t know the ECB or BOJ notes well enough to say much there, but it doesn’t really matter, because again, trivial to prove their missing low is a choice — is there nothing they can do to miss high if they choose to? Are there no assets in the world left to purchase? Do they no longer own printing presses? 🙂
Or see this interview:
http://new.dowjones.com/newsletter/grand-central-unemployment-falls-fed-doves-pivot-low-inflation-concern/
“I think it is a signal of underutilization of resources,” Mr. Kocherlakota said. An inflation-targeting central bank should be as worried about undershooting inflation as well as overshooting it, he said. With labor market measures like the unemployment rate sending mixed and complicated signals about the economy, debate at the Fed in the months ahead could turn to inflation, and how long the central bank is willing to tolerate undershooting its target.
So by implication, they aren’t. The Fed itself says the Fed is purposely missing low.
Never use an empirical model when you can just ask a person why they did something.
* they arent [as worried about undershooting inflation as well as overshooting it]
By how much are they choosing to miss low? Is it exactly the difference between the observed rate and the target? Magic!
Let me update the logic above:
P = inflation is x%
Q’ = CB is targeting (x+δ)% – δ% inflation (target inflation x+δ, and missing low by δ)
Axiom P→Q’ (assume a CB can always get x% inflation)
Observe P (inflation is x%)
Therefore Q’ (CB is targeting (x+δ)% – δ% = x% inflation)
Obviously the Fed cannot exactly target inflation with arbitrary precision. That’s as silly an argument as supposing they have no printing presses.
In case anyone’s interested, Jason continued this with a post here:
http://informationtransfereconomics.blogspot.com/2014/06/how-does-liquidity-trap-work.html