Monetary offset refutation, or CYA?
I can’t say I am surprised by what people are writing. But it’s still disappointing. Let’s review the past 6 months:
1. The Fed sees fiscal austerity on the horizon in late 2012, and tries to offset it with several monetary initiatives. They even say they are doing the stimulus with the intention of shielding the economy from austerity. Monetary offset.
2. The fiscal austerity that the Fed was worried about does happen.
3. The Fed forecasts 3% to 3.5% RGDP growth over the next couple of years, knowing full well that fiscal austerity was on the way.
4. Over here at TheMoneyIllusion I am skeptical that we’ll get 3% to 3.5% RGDP growth, based on market indicators. I claim that the Fed actions are a modest plus (again based on market reactions) but anticipate that it will probably just roughly offset the fiscal austerity, leaving RGDP and NGDP growth in 2013 about the same as 2012.
5. So far it looks like I was right and the Fed was wrong. Admittedly things could change, but Fed officials are already in full CYA-mode, blaming fiscal austerity for the slowdown, despite the fact that they anticipated fiscal austerity when they made their forecasts in late 2012. You don’t hear them say; “Sumner was right, we never should have forecast such robust RGDP growth in the first place.”
6. Meanwhile the anti-austerians predicted a 1.5% hit to RGDP, some even mentioned 2%, and yet those predictions also look off base so far, as 2013 will look much like 2012. Yet the press has an overwhelming Keynesian bias, so any facts are assumed to support the Keynesian model, no matter how much at variance. For instance, the US is doing more fiscal austerity that Europe, and yet David Beckworth showed that it’s Europe that has the big slowdown, not the US. And let’s not even talk about the fact that the biggest surge in public debt in global history—-Japan since 1993, has been associated with FALLING NGDP!!!! That’s right, the biggest surge in public debt is associated with the worst 20-year performance for AD in all of global history. Of course “correlation doesn’t prove causation,” unless it supports the Keynesian model.
Karl Smith quotes Jeremy Stein, showing that it takes a PhD in economics from an elite school to concoct a theory that a central bank with a monopoly on the printing of money is somehow unable to debase its own currency:
Moreover, . . , absent policy action, progress on reducing unemployment will likely be slow for some time. Meanwhile, inflation is subdued, running at or below our long-run objective of 2 percent, while inflation expectations remain well anchored. If the federal funds rate were at, say, 3 percent, we would have, in my view, an open-and-shut case for reducing it.
The complication, of course, is that the federal funds rate is essentially at its lower bound, which means that we cannot do more simply by turning that dial further. Instead, we have to use unconventional tools, such as LSAPs and guidance about the future path of the federal funds rate.
OK not unable, just afraid that a policy that produces a million new jobs for the desperately unemployed might lead to a trillion dollar capital gain for the Federal government. Why would that be a problem? Because it might be composed of a $200 billion loss to the Fed, and a $1.2 trillion gain to the Treasury, and we all know that the Fed is an independent part of the government, which must never, ever incur a loss, no matter how vast its profits have been in recent years. (Yes, I’m being sarcastic.)
Here’s what Stein is actually saying, if you translate:
We are going to allow a million workers to remain unemployed, because we are too timid to switch from growth rate targeting to level targeting.
This requires some explanation. Under growth rate targeting it’s perfectly possible to miss your implied NGDP target of 5.5% by 1.5% each year, leading to 4.0% growth, year after year. That’s not far from what’s happened since 2009.
Under level targeting (5.5% trend line) if you miss your target by 1.5% each year, the actual NGDP growth rate will still be exactly 5.5%, in all but the first year. (Think about it.)
Why is this crucial? Because the demand for base money is negatively related to the NGDP trend growth rate. So under NGDPLT, a more expansionary policy (higher target NGDP growth) actually leads to less monetary base as a share of QE (Australia’s a good example here) and thus less of that QE that the Fed is so afraid of doing.
Here’s the great irony; the Fed thinks easy money is too risky, and yet the things it (falsely) thinks are risky about easy money (a bloated base) are actually associated with the low NGDP policies of the US and Japan, not the high trend NGDP policy of Australia (6.5% per year since 1996.)
When I think of all the damage being done by these conceptual errors it makes me want to cry.
PS. Leaving “rant mode” for a minute; I do concede the Fed didn’t forecast 100% of the recent sequester, and hence Q2 NGDP growth will likely fall a bit below “Sumner Critique” predictions. I address that issue in this recent post, which some commenters seemed to misinterpret as support for fiscal stimulus.
PPS. Why is the Fed afraid of level targeting? Because it’s a powerful tool. Here’s an analogy. I’ve got gangrene throughout my left leg. I ask the doctor for advice. He suggests a large axe. I say “how about some drug therapy instead?” Why am I afraid of the axe? Because it really will work, it really will get rid of my gangrene leg.
PPPS. Nick Rowe has a great post, which is loosely related to the “monetary stimulus is risky” idea that I’ve tried to refute here. Ryan Avent also has an excellent post on a similar theme.
PPPPS. Is there anyone out there who has drunk so much Keynesian Koolaid that they think that, in the absence of fiscal austerity, in late 2012 the Fed would have been predicting 5% RGDP growth going forward?
Tags:
2. May 2013 at 18:13
When I think of all the damage being done by these conceptual errors it makes me want to cry. –Scott Sumner.
Well ,cry me a river, I cried a river too (great song, btw).
What happened to the economics profession?
Now, the 1960s are dismissed as “having caused inflation.” Inflation was five percent when the decade ended, and real incomes per capita were up 30 percent in 10 years.
But Volcker “beat inflation.” It was five percent when he finished.
Now, economists are dementedly, peevishly fixated on rates of inflation like 1 percent or 2 percent. Nirvana is perfect 0, or maybe (priapism time) even mild deflation, obtained by productivity gains.
The Fed says it might “ease” and allow 2.25 percent.
And yes, you want to build national debts? Try tight money.
Like Sumner says, you gotta have a Phd from the best school in the world to think like this.
2. May 2013 at 18:14
Did the Fed have specific projections regarding the sequester and the fiscal cliff?
2. May 2013 at 18:37
“Because it might be composed of a $200 billion loss to the Fed, and a $1.2 trillion gain to the Treasury, and we all know that the Fed is an independent part of the government, which must never, ever incur a loss, no matter how vast its profits have been in recent years. (Yes, I’m being sarcastic.)”
This does seem to be the real issue. No one really doubts that highly aggressive Fed action could engineer NGDP growth, but the question is what effect it has on economic incentives when doing so involves permanently monetizing debts. Although I would suggest that the effect on incentives is to a large extent retroactive, which is to say a lot of market participants have (implicitly?) acted in the anticipation of such monetization occurring and now depend on its follow-through for their financial survival.
However, I think it is possible, theoretically, for the Fed to get the effects of permanent monetization by employing solely forward communicative methods. As soon as the market considers certain debts “for all intents and purposes monetized”, the effect is already delivered.
Would the effect on incentives be the same as that of true, direct permanent monetization? I tentatively think the answer is yes. Which means that probably much of what has been going on economically in the last few decades (finance sector growth, rising inequality, wage growth not matching productivity growth, etc) has to be understood in these terms.
ps. I arrive at the conclusion that permanent monetization is needed simply by postulating (with reason) that Western economies are loaded down with private debt resulting from past asset bubbles, and this debt confers disinflationary pressure on the respective consumption economies (see for example: http://michael-hudson.com/2012/09/incorporating-the-rentier-sectors-into-a-financial-model-3/).
2. May 2013 at 20:25
Michael Hudson:
I agree with permanent, or at least long-term (permanent is a long time), monetization of the federal debt (I would not do QE on mortgage debt, will lead to politicization of the prices).
But the paper you cite and wrote is nearly impenetrable! Consider a translation into English. If you need help, let me know.
2. May 2013 at 20:44
Demand for commodities is not demand for labor. Increasing spending on final goods does not increase the demand for labor one penny.
The monetary base is in fact risky, because it ultimately (via the operational constraint of capital reserve ratios) constrains the amount of bank credit inflation that can take place. The base has rapidly grown since 2008. It contains a very high potential of inflation, and if the Fed later stops this through contraction, then it will bring about another correction.
There is a difference between base in absolute terms, and a base relative to NGDP. Australia doesn’t have a small base on the basis that it has stable NGDP, and the US doesn’t have a large base on the basis that it has unstable NGDP. The base is fully controlled by the central bank. The market isn’t producing base money. If the base is low (high), it’s because the central bank wanted it that way. It isn’t deterministically caused by NGDP.
3. May 2013 at 01:50
Off-topic, but I just read about something called “Macon Money” (http://www.knightfoundation.org/macon-money-game-evaluation-summary/)
It was designed in a completely different context, but I think can have big monetary implications, if redesigned the right way. My idea is a simpler version of Miles Kimball’s e-money. I have it here (http://ashokarao.com/2013/05/03/maconomics/) but here’s the TLDR since you are busy:
What if parity conditions were relaxed and exchange rate with the dollar was governed by market forces and the Macon Mint? This policy would require three guiding conditions:
1. Macon Money will never appreciate over parity on the dollar.
2. Severe depreciation would be binding. That is, if Macon Money falls below any $n ∈ N, where N is a predetermined set of values (including the floor) this becomes an effective cap on its value.
3. Some determined time after the Macon falls to F, it will no longer be guaranteed against the dollar.
The trick here is creating a unit of exchange that is in the public conscious not a store of value. This gives the government broad scope to inflate that currency to spur consumption, without hurting savers. The dollar would loose some value, as the Fed guarantees an exchange rate. Today some people are saying we need 4% inflation or more to stabilize AD. This would be a cheaper way to do the same thing. *And because it’s not a store of value, the Fed can use some opaque stochastic techniques to change the value of Macons overcoming the rational expectations critique. This is important.* (I describe this more in the post.)
There are also some behavioral factors I just thought about. A lot of people (in the study) enjoyed using it because of the gamification element. This decreases chances of rapid sell-off. And people may not care much about depreciation against the dollar because they got it from nothing.
At worst, people just immediately trade the currency for dollars. This reduces to a plain monetary stimulus of the helicopter sort. On the other hand, it can support a sustained increase in money velocity, commerce, and AD recovery. And if things ever “heat up” too much the Fed can just devalue the currency. Because it came out of nowhere.
3. May 2013 at 01:57
Somewhat more on topic, I think a NGDPT regime today would cause more inflation. I think both CBO and Fed are overestimating the output gap. This means we’d hit full employment sooner than expected, requiring inflation to prop up necessary nominal growth.
This is evidenced by the serious overestimates of RGDP growth. By the way, you note that a more optimal indicator might be GDP less indirect taxes. In the same spirit might we also discount our capital consumption allowance?
3. May 2013 at 02:34
Scott,
here in Germany (in the heart of darkness) an often used reply to Market Monetarists goes like this:
Monetary policy at its worst does nothing, at its second best it could raise asset prices but not NGDP and at best it raises NGDP through the wealth effect.
All of this is bad. In the first case, central banks push money into the markets buying government bonds. This will do nothing, money demand rises and the new money just sits there. In the second case, people will take the money and bid up asset prices. When people feel that asset prices won’t go higher, money demand rises and the money just sits there. In the best case, rising asset prices will make people feel richer, so they start consuming more, raising NGDP. But even this best case scenario is bad for the economy, because it’s just a new asset bubble creating the illusion of wealth luring people into consumption instead of doing some real investments which would make us richer in real terms.
In short, NGDPL targeting won’t work and even if it does work, it’s bad.
The “won’t work” part is easy to counter, but what would you answer to the “new asset bubble” scaremongering? Like in this new Roubini piece:
http://www.project-syndicate.org/commentary/the-federal-reserve-s-policy-dilemma-by-nouriel-roubini
3. May 2013 at 03:21
Nice rant.
3. May 2013 at 04:58
Geoff wrote:
“There is a difference between base in absolute terms, and a base relative to NGDP. Australia doesn’t have a small base on the basis that it has stable NGDP, and the US doesn’t have a large base on the basis that it has unstable NGDP. The base is fully controlled by the central bank. The market isn’t producing base money. If the base is low (high), it’s because the central bank wanted it that way. It isn’t deterministically caused by NGDP.”
I don’t think Scott’s argument is that the base is deterministically controlled by NGDP. (As you suggest, the base is deterministically controlled by the Fed). An environment of low NDGP growth will lead the Fed to react by expaning the base.
A five alarm fire does not deterministically control the flow of water into the burning building, but it will lead the fire department to do so.
3. May 2013 at 05:07
Aidan, Not sure about the sequester, but they surely know about the cliff.
Rademaker, In my view monetization would lead to hyperinflation.
Ashok, Interesting idea, but nothing so radical is needed.
Yes, we could dump both depreciation and indirect business taxes, which I believe leads to national income. That’s probably better than NGDP.
libertaer, In the land of the 1923 hyperinflation they don’t believe that printing lots of currency is inflationary? Are you sure? I thought the Germans opposed monetary stimulus precisely because they feared higher inflation.
3. May 2013 at 05:09
Michael, Actually I am talking about real base demand, although the innovation of IOR does change the picture somewhat, boosting base demand higher than NGDP growth alone would predict.
3. May 2013 at 05:32
libertaer wrote:
“Monetary policy at its worst does nothing, at its second best it could raise asset prices but not NGDP and at best it raises NGDP through the wealth effect.”
Libertaer, I think you are conflating NGDP and RGDP. A central bank can raise NGDP via inflationany time it chooses to do so. It cannot necessarily raise RDGP – that would depend on the situation (e.g. is the economy at potential or below potential).
3. May 2013 at 06:20
Scott – I meant in their projections, what were they assuming policy would be? Did they assume X amount of tax increases and spending cuts from the sequester? Did their assumptions include sequester being fully/partially repealed, and the same with the Bush tax cuts?
3. May 2013 at 06:33
Who predicted a 1.5% hit to RGDP? In any case you don’t need that to happen to justify opposing austerity.
Your admission that “Some have argued that fiscal austerity is slowing the recovery. Indeed the Fed has argued that fiscal austerity is slowing the recovery. And that’s because fiscal austerity is slowing the recovery. RGDP growth in Q2 will likely be lower than if the sequester had not taken effect. So why do I keep prattling on about “zero fiscal multiplier?”
http://www.themoneyillusion.com/?p=20971
Is all anyone needs to argue against the sequester.
3. May 2013 at 06:42
“the US is doing more fiscal austerity that Europe,”
On what basis do you make that claim? It’s the first I’ve heard it.
3. May 2013 at 07:22
“libertaer, In the land of the 1923 hyperinflation they don’t believe that printing lots of currency is inflationary? Are you sure? I thought the Germans opposed monetary stimulus precisely because they feared higher inflation.”
Very often the same guys who -just a second ago- feared hyperinflation are the first ones to argue that money creation won’t work. I think it’s obvious why.
But what’s your reply to “asset bubble” scaremongering?
3. May 2013 at 07:29
Michael,
“Libertaer, I think you are conflating NGDP and RGDP. A central bank can raise NGDP via inflationany time it chooses to do so. It cannot necessarily raise RDGP – that would depend on the situation (e.g. is the economy at potential or below potential).”
I agree. But the people who fear asset bubbles argue that monetary policy will hurt RGDP by creating (through rising asset prices) an illusion of wealth, tricking people into consumption instead of investment.
Long ago Nick Rowe did a post on this: http://worthwhile.typepad.com/worthwhile_canadian_initi/2009/08/the-grain-of-truth-in-the-junker-fallacy.html
3. May 2013 at 07:36
On “risky monetary stimulus”:
http://thefaintofheart.wordpress.com/2013/05/02/effective-x-ineffective-tools/
3. May 2013 at 09:02
If the FED had the will fiscal policy wouldn’t matter as much right now (until full employment).
Japan had the biggest shrinking GDP in modern history although I would guess that there were some consistent supply shocks like drought and disease. The black death (Chinas first great export) prevented gdp growth for centuries.
3. May 2013 at 09:24
Scott you said, “Ashok, Interesting idea, but nothing so radical is needed.”
Is it so radical for the Fed to create a temporary currency? And it seems to create “social capital” and “friendships” too. That seems nice. In fact, this might be a way to go about seriously expansionary monetary policy without inflating an asset bubble (I have qualms with this argument, but that’s beside the point).
“Yes, we could dump both depreciation and indirect business taxes, which I believe leads to national income. That’s probably better than NGDP.”
I think this makes a big difference with the baby boomer generation. I actually have no idea about the technical accounting of depreciation and capital consumption etc. But when a mass wave of (relatively educated) people retire we’re loosing tons of human capital, meaning a lot of education is just replacing that.
So I think this might be a lot more important in the future.
3. May 2013 at 10:00
[…] – Scott Sumner […]
3. May 2013 at 12:27
Aidan, I don’t know.
Mike Sax, You said;
Who predicted a 1.5% hit to RGDP?”
Just about every Keynesian article I read. Unless they predicted a 2.0% hit.
You said;
“Is all anyone needs to argue against the sequester.”
I’m afraid you didn’t understand the post, read it again.
The estimates I’ve seen show the US doing slightly more austerity than Europe this year, I don’t recall the source.
libertaer, Asset bubbles don’t exist, and if they did they would not matter. That’s my response.
errorr, Yes, I meant modern history.
3. May 2013 at 12:54
Well isn’t that pretty key when assessing how well the Fed’s projections are doing? You “concede the Fed didn’t forecast 100% of the recent sequester,” but there’s a lot of space between 0% and 99%.
3. May 2013 at 18:23
Aidan, I don’t think you can put a number on these things, as each FOMC member has his or her own view.
8. May 2013 at 05:07
[…] Here’s a post on monetary offset. Read point 6 in particular. Fiscal austerity is deeper in the US than the Eurozone, but monetary policy is easier. US is growing faster. Money wins. http://www.themoneyillusion.com/?p=21008 […]
31. May 2013 at 04:30
Perhaps the Fed is right that easy money is too risk, since it will likely lead to more reckless goverment spending.
30. July 2013 at 18:21
[…] in April I predicted a weak 2nd quarter RGDP number, mostly due to the sequester. Tomorrow morning we will find out […]