dtoh on DeLong
Dtoh recently left this comment after my DeLong post:
“Believers that only fiscal policy matters at the zero lower bound …. [are] wrong in [their] belief that monetary policy is pushing on a string at the zero lower bound.”
Seems to me that statement completely cuts PK off at the knees.
It acknowledges both a) that fiscal policy is not necessary at the ZLB and b) that there is a monetary offset.
Don’t think you could ask for much more.
Oh I could ask for much more:
I could ask for DeLong to correctly describe market monetarist views on monetary offset. Views he heard me describe while he sat next to me at a bloggers conference.
I could ask DeLong not to suggest that MMs believe monetary policy works by changing long term interest rates.
I could ask DeLong to correctly describe MM views on temporary vs. permanent monetary injections.
I could ask DeLong to point out that Krugman was being misleading when he implied that market monetarists claimed 2013 was a test of MM, whereas it was Krugman himself who made that claim.
I could ask DeLong not to describe me as being “deranged.”
I could ask DeLong not to describe me as “losing my mind.”
I could ask DeLong not to suggest that I “don’t know how to read.”
But then it would be no fun reading his blog.
Tags:
7. January 2014 at 07:33
Scott,
Here’s what I’ll ask: provide the appropriate descriptions of your view that you think DeLong got wrong.
Specifically:
1) I still don’t understand your use of the concept of monetary offset. Please explain.
2) Is monetary policy effective through hot potato effect instead of long term interest rates?
7. January 2014 at 07:42
John: Maybe in the meantime before Scott replys (if he does) both answers can be easily found using the search function on this page.
1) Searching for “monetary offset” returns a blogpost also linking Scott’s paper for the topic: http://www.themoneyillusion.com/?p=23559
2) If searching for “transmission mechanism” the very first result is for post that in turn links to a very good post by Nick Rowe: http://www.themoneyillusion.com/?p=24433
7. January 2014 at 07:49
Ah ha, DeKrugman sticks up his head!
Look, rather than argue with them, why not ask them if they are onboard with pure corporate, payroll, and SMB income tax cuts as “Fiscal Stimulus”
Are they willing to tell the Democrats that the unemployment problem is such a crisis that deficit finances tax cuts for the top half are worth it?
Make them admit, there is a thing they don’t love the most, and their argument with MM falls apart.
7. January 2014 at 08:08
John, This is monetary offset:
http://mercatus.org/sites/default/files/Sumner_FiscalMultiplier_MOP_090313.pdf
And yes, it’s the HPE, we don’t even assume that long term rates fall with stimulus—they didn’t in the 1960s and 1970s
7. January 2014 at 09:10
Scott, DeLong’s post does not seem to have grasped reality. Incorrectly describing your views and Krugman’s “challenge” does not really advance the ball.
I would note that in the mercatus paper you link you write “In the United States, Congress passed a nearly $800 billion stimulus in early 2009, yet growth remained sluggish.” This is a bit too reminiscent of the claims of some of your critics that the Fed has done a whole lot, yet growth remains sluggish. In both cases, the answer seems to be, yes but they didn’t do nearly enough.
7. January 2014 at 09:24
It’s all about the baseline
http://www.thefiscaltimes.com/Columns/2014/01/07/How-Tell-If-Fiscal-Policy-Works
7. January 2014 at 09:24
“I could ask DeLong not to describe me as being “deranged.”
“I could ask DeLong not to describe me as “losing my mind.”
“I could ask DeLong not to suggest that I “don’t know how to read.”
Scott has said the same things and worse about me. I wonder why he finds it so egregious if Delong does it.
7. January 2014 at 16:29
Mike Sax:
[citation needed]. Those certainly don’t sound like things Scott Sumner would say. But I admit I don’t always read your entire comments or Scott’s replies to them; I await your links.
7. January 2014 at 16:41
foosion: baseline, yes, but hasn’t that been Scott’s point from the beginning (the “Sumner critique”)? It makes no sense to try to analyze the fiscal multiplier, without having a model of the monetary policy reaction function.
7. January 2014 at 16:49
William if you have any doubt what I’m saying you haven’t been reading here very long. I mean I would say check the archives in almost any post I left comments in.
You want links start here.
http://diaryofarepublicanhater.blogspot.com/2013/10/why-sumner-claims-im-ignorant-of.html
http://diaryofarepublicanhater.blogspot.com/2013/11/how-many-nails-does-it-take-to-bury.html
7. January 2014 at 19:12
ssumner
I looked at the intro course on money but couldnt find anything on the transmission mechanism. Sorry could you direct me more specifically to the part regarding the transmission mechanism. What is your definition of the transmission mechanism?
7. January 2014 at 19:41
Scott,
As a liberal who appreciates much of what Krugman does, and even more what he tries to do, it’s clear to me that market monetarism wins hands down. The Keynesians have been totally discredited to anyone not blind with that theoretical ideology.
Does that mean market monetarism is correct? No, but my best guess is that it is, or is very close to being correct. It’s hard to see how it could be much clearer.
7. January 2014 at 20:52
lxd,
Look at the link in JV Dubois’ comment above. There is a good discussion there. Also Scott has in the past pointed to textbooks on Monetary Econcomics by Mishkin. You can find the entire text for older editions online at no cost.
That said, in my opinion, MMs have the transmission mechanism slightly wrong…. mostly because their thinking is anchored in various forms of earlier QTM orthodoxy.
7. January 2014 at 22:19
dtoh
SSumner assumes that when the base is increased (exogenously and permanently) the public’s preferred MB/GDP ratio stays the same. He is assuming away the problem because the transmission mechanism determines if base will increase endogenously and permanently. For example if the transmission mechanism is commercial banks dealing with the fed or other entities with a low MPC then expectation s wont budge. If base is expanded to entities with a high MPC like the general public then more money will be used to consume as opposed to portfolio re-balancing which in turn will lift ngdp expectations.
8. January 2014 at 02:48
Scott Freeland-then you’re not really a liberal. The whole point of MM is that the GOP is never wrong no matter how much it cuts the budget by-as the Fed will offset it.
I mean on economics at least what liberal positions do you actually have left? Sumner’s whole model is that the Fed deals with demand side issues and the fiscal authorities should do supply side stuff-tax cuts for the rich, cut programs for the poor and lots of deregulation.
This is the beauty of MM-if you’re a conservative. This is why Morgan Warstler is right to love MM. However, you’re just fooling yourself if you think that you can really be a MM liberal.
8. January 2014 at 04:46
Mike Sax,
I think you’re failing the ideological Turing test in that first paragraph.
Also, equating liberalism (in the American sense, or indeed any other sense) with Keynesianism is bizarre. Firstly, because Keynesianism is hardly the only efficiency argument that can be put forward for government interventions in an economy; secondly, because efficiency arguments are hardly the only arguments that can be put forward for government interventions.
Only if you think that conservative arguments for those supply-side measures are microeconomically and morally unassailable does your conclusion follow, which is a puzzling position for an alleged Republican-hater to take.
8. January 2014 at 05:53
foosion, You said;
“This is a bit too reminiscent of the claims of some of your critics that the Fed has done a whole lot, yet growth remains sluggish. In both cases, the answer seems to be, yes but they didn’t do nearly enough.”
That’s true, but here’s the crucial difference. Fiscal stimulus is costly whereas monetary stimulus is not. Indeed greater monetary stimulus would not even require the Fed to buy more securities. It’s easy to double NGDP with monetary stimulus. If you tried to do it with fiscal stimulus you’d bankrupt the government.
Mike, You said;
“Scott has said the same things and worse about me. I wonder why he finds it so egregious if Delong does it.”
I don’t find you deranged, but I do think you are unable to read–for instance the last line of my post.
You said;
“The whole point of MM is that the GOP is never wrong no matter how much it cuts the budget by-as the Fed will offset it.”
This is unbelievably stupid even by your standards. I’m not even a Republican. And the GOP increased the budget sharply last time they were in power–or have you already forgotten.
lxdr, The basic transmission mechanism is the hot potato effect, it is covered in the course. Also the expected future hot potato effect.
8. January 2014 at 08:52
And you can certainly be a MM liberal, but you’d distinguish yourself as such by concerning yourself with efficient government action.
#1 Imagine The Financial Crisis as as a terribly time for conservatives bc the “market” went off the rails and making people afraid and desperate for govt. action.
#2 Imagine Obamacare as the exact opposite for progressives.
MM can solve for #1, that’s why I like it (along with some of your assertions).
MM makes the problems of #2 more obvious. So you’d need to be a strong neo-liberal efficient government guy to cheer it.
8. January 2014 at 12:27
Civility is valued when the intent is to communicate.
BDL/Krugman’s writing style is both a sop to and a reflection of their audience, which skews young and progressive. When Krugman wrote like an economist, he was just another mildly interesting Nobel winner; writing as the left’s Ann Coulter he’s garnered a far larger audience.
8. January 2014 at 18:04
lxdr,
Think of it this this way. If the ultimate counter-party to OMP is the non-banking sector, then there is a decrease in net financial assets held by the non-banking sector (e.g. draw-down on a line of credit, higher credit card balance, etc), it will generally result in an increase in AD because the non-banking sector is exchanging these financial asset in order to purchase goods and services. If on the other hand the ultimate counter-party to OMP is just a bank, then all that happens is that bank reserves go up and there is no effect.
I think you’re saying the same thing, just in a slightly different manner…. although to use your description, you should probably say marginal propensity to spend rather than consume since the effect of OMP includs both consumption and investment so “spend” is probably more inclusive and accurate.
8. January 2014 at 22:23
dtoh
“If the ultimate counter-party to OMP is the non-banking sector, then there is a decrease in net financial assets held by the non-banking sector ”
If we assume that the CB conducts policy through purchases of assets such as treasuries or other assets I agree. But when conducting policy directly with the public the CB does not need to purchase anything and it can simply increase balances of recipients which will increase net financial assets. The CB can recognize new created base as equity (bearer)instead of liab in order to not undermine balance sheet.
I agree about MPS as a better term to use although consumption does account for about 65-70% of gdp.
8. January 2014 at 22:57
lxdr,
Yes if the Fed were doing “helicopter drops” the mechanism would just be HPE.
Consumption is big chunk of GDP, but compared to investment it’s less influenced by changes in rates and expected NGDP.
10. January 2014 at 02:11
dtoh
Under OMO’s or QE the HPE is limited in that expectations wont budge much though. Expectations of inflation or ngdp will be affected much more by heli drops. The counterparties to the fed are vital.
If a bunch of banks or high wealth individuals/companies receive relatively more base compared to other assets the market will not expect much higher ngdp or inflation if lending markets are saturated and assets arent widely held. For example over 90% of stocks are held by the top 20%, so if the wealthy experience greater it is not expected they will spend much more.
10. January 2014 at 02:44
lxdr,
I’m not sure I’m following you, but a couple of points.
1. OMO and QE are the same thing. Just different names. Either way it’s the Fed exchanging money for financial assets.
2. The Fed doesn’t do heli drops so there is no real points in talking about them.
3. Nobody “receives” base. It’s always an exchange of financial assets for money (base). (A lot of the bad thinking about monetary policy arises because people think of an increase in the base as an exogenous event…like an increase in the apple crop of a new gold discovery.)
4. When non-banking sector market participants enter into this exchange it’s because they want to exchange financial assets for goods and services. (Money is merely the intermediate medium of exchange for doing this.) So it’s self selecting. The only people who enter into the exchange have an MPS (to use your term) of very close to 100%.
5. I disagree with Scott on the HPE. He thinks it’s causal. I think it’s a good model, but does not explain causality.
10. January 2014 at 04:13
dtoh
1. I know.
2. The fed doesnt do heli drops but in order to make monetary policy effective it should IMO.
3. By receive I meant receive in exchange for treasuries or MBS or whatever, I wasnt saying that some entities just get new money for nothing.
4. That’s why under heli drops increasing the general public’s money will result in higher ngdp and inflation expectations when compared to omo’S or QE. People will exchange their money balances for goods and services not just hold excess reserves or cash balances.
5. I can see the reasoning behind hpe on some level but “permanently” increasing base isnt exogenously determined under current system and the price level wont increase in lock step with the increase in base as he assumes in his writings.
10. January 2014 at 13:48
lxdr,
Briefly on why heli drops won’t work.
1. You need monetary policy because you have sticky wages and prices. When you have a shock (financial asset prices drop suddenly), there is a reduced exchange of financial assets for real goods and services… in other words AD drops. With sticky wages and prices, the market can’t adjust quickly and you get unemployment.
2. Monetary policy works by causing a marginal increase in the exchange of financial assets for goods and services either by raising the price of financial assets relative to goods and services or by raising expected future NGDP. If the price of both financial assets and goods/services rose simultaneously and at the same rate, there would be no increased inducement to exchange financial assets for goods/services (i.e. no increase in AD).
3. A helicopter drop might work once or twice. However, once the public/market figured it out, the prices of both financial assets and goods/services would rise simultaneously and instantly whenever the Fed issued new base; because the relative price of financial assets and goods/services does not change, there would be no marginal increase in the exchange of financial assets for goods and services; and no increase in AD.
10. January 2014 at 15:39
1. If you have a shock which results in a reduced exchange of financial assets for goods and services the CB can respond by increasing peoples money balances. This means people will have more disposable cash and will therefore spend more which will offset the AD shortfall.
2. Increasing everybody’s spending capacity through “heli drops” will increase the relative price of financial assets because valuations will increase from expected higher ngdp. Higher demand will increase valuations of stocks for example. Higher expected ngdp will occur because it is expected that heli drops will result in more spending.
3. Heli drops will work every time they result in expected higher spending. If people have more money they will always spend more to some degree ceteris paribus. So long as the fed conducts heli drops within the parameters of an inflation or ngdp target heli drops will remain credible and will continue to affect expectations.
Prices for goods and services wont change in lockstep with changes in financial assets prices. There is no mechanism for this. prices for goods and services may go up faster or slower than financial asset prices. Price setters dont set prices entirely based on financial markets. The price of goods and services is determined by many factors.
10. January 2014 at 19:34
lxdr
If the Fed does regular and predictable heli drops then the market will figure it out and you will just get instantaneous jumps in all prices with no effect on real demand or output.
The reason OMP works is because the Fed is buying one particular thing (financial assets), which raises their price relative to the price of goods/service so you get an exchange and an increase in real assets.
With heli drops the price of everything goes up and nothing else happens.
10. January 2014 at 21:48
dtoh
Prices dont move 1 for 1 along with increases in base. Price setters dont monitor base in order to set prices.
If you have slack labor markets, excess capacity etc… an increase in demand from heli drops wont translate into higher prices or inflation expectations. As a result you get increases in real gdp.
Base never increases at the rate of CPI increases historically in the US for example.
10. January 2014 at 22:12
lxdr,
I have to respectfully disagree. Base historically has not moved directly with with CPI because the US has never done heli drops. If you look at countries where they have done heli drops (i.e. actually just printed money like Zimbabwe, Argentina, etc,), prices get adjusted hourly. Market participants are rational. If everyone suddenly has 10% more dollars, prices will rise by 10% almost immediately. Perhaps not the first time the Fed were to do a heli drop, but on subsequent drops that’s exactly what would happen.
10. January 2014 at 22:26
“Market participants are rational. If everyone suddenly has 10% more dollars, prices will rise by 10% almost immediately.”
Yes they are rational. If people have 1% more money expanded into their accounts over a 3 month period in weekly increments it is rational to expect a 1% increase in AD. Whether that translates into higher inflation depends on several factors like is the economy at potential output. If their is slack in labor markets as well as excess capacity their is no reason to rationaly expect a significant increase in inflation.
If the fed managed “heli drops” in an imprudent manner then credibility would be lost and I agree with you or maybe you would get higher inflation than rate of base increase like ARG or ZIM.
Rate of heli drops needs to be managed under inflation targeting or ngdp targeting regime not just ad hoc zimbabwe system.
10. January 2014 at 23:32
lxdr,
I still don’t think that gets to the root of the problem and this is fundamental to a lot of the discussion about the transmission mechanism. Scott and a lot of others grounded in QTM, think of increases in the base as an exogenous event (like a bumper apple crop if apples are the medium of exchange) and thus they conclude that the mechanism is the HPE. If people have more money they will spend it.
But that’s not what’s happening. The reason you get the business cycle, and the reason you need monetary policy and the very reason that monetary policy is effective is because of changes in the price of financial assets relative to the price of goods and services. Shocks cause a drop in financial asset prices and a reduction in AD. Fed action raises financial asset prices and thereby increases AD because there is a marginal increase in the exchange of financial assets for good and services resulting from the change in relative prices.
11. January 2014 at 00:20
“If people have more money they will spend it.”
Ceteris paribus I think so too. They will spend some proportion of it whether that be 80% or 20% is the only question. What they don’t spend will be used to pay off debt or just generally improve their balance sheet making them more eligible for credit.
Increasing money balances of people will increase financial asset prices because of the greater spending by people. Financial asset prices such as stocks are determined to a large extent by expectations of how underlying companies will perform in terms of sales and profits. Greater spending means greater income for companies.
Don’t only reason from a price change. What about quantity? Increasing the supply of a financial asset such as money will bring down its price but its increased quantity will
result in a marginal increase in the exchange of financial assets for good and services resulting from the change in relative quantities.
11. January 2014 at 10:07
lxdr,
I generally agree with what you say.
Just so we’re clear on nomenclature. When I say “financial assets,” I’m distinguishing from money. If I talk about buying or selling “financial assets,” I’m generally referring to any increase in net (long or short) holdings of financial assets so to me paying off debt is “buying” a financial asset. When I talk about the real price of financial assets, I’m talking about 1/the real (after inflation) risk adjusted IRR so if expected sales and profits go up, the nominal price of a stock may change but it’s real price after the nominal price adjustment will stay the same. The real price of a stock would only change if the expected risk adjusted after tax IRR changed.
Coming back to the question of helicopter drops. Suppose that the Fed announced that on January 1, 2015, “Current Dollars” would expire and that before then everyone will need to exchange Current Dollars for New Dollars at the rate of 2 New Dollars for every 1 Current Dollar. Excluding the cost of trips to the bank, etc., would that have an impact on real spending and real output? I think the answer is that it would have no impact. So if the Fed is doing helicopter drops, then functionally it should be identical to exchanging Current Dollars for New Dollars. (I assume with a helicopter drop this is what you’re talking about… i.e. you get extra dollars in proportion to your current dollar holdings.)
You talked earlier about doing helicopter drops in a prudent or credible manner. I don’t think it actually has anything to with prudence or credibility… only transparency. I have great faith in the rationality of markets and economic participants so I believe that if the Fed is doing helicopter drops in a transparent fashion regardless of how prudent they are, then for the most part they will cause a price effect but no real effect.
11. January 2014 at 17:04
“Suppose that the Fed announced that on January 1, 2015, “Current Dollars” would expire and that before then everyone will need to exchange Current Dollars for New Dollars at the rate of 2 New Dollars for every 1 Current Dollar.”
This would have no effect. Its not an expansion of base its a change of currency from current dollars to new dollars. Everyone exchanges one currency for another and reports prices in the new currency. Through heli drops more of the existing currency enters the system which will increase demand and through demand prices may be affected depending on how large, frequent heli drops are.
Why would an expansion of base through heli drops cause a price effect exactly proportional to the expansion in base without increasing real growth? What is the mechanism for this?
An expansion of base hasnt historically been matched by inflation. There should be no difference in the outcome if base is expanded through heli drops or through OMO’s if its done under inflation targeting for example. Why is the effect different on prices or growth if base grows through OMO’s or heli drops?
11. January 2014 at 21:58
lxdr,
Suppose the Fed didn’t issue a new currency. Suppose they just said anybody could take cash into any Fed member bank on the First of January and exchange one green bill for two red bills, or one bill for two bills with a picture of different dead President, or suppose you hand in one old green bill and get two green new bills in exchange. Or suppose the bank just recorded the serial number (so the bill could only be used once) and gave the holder one more dollar. Or suppose they gave you a quarter. It’s all the same.
The reason OMP work to raise real AD is because the Fed doesn’t do helicopter drops. They exchange money for financial assets. This raises the price of financial assets relative to goods and services, which causes a marginal increase in the exchange of financial assets for goods and services. Manufacturers draw down credit lines to build factors, consumers take out home loans to buy new houses, etc., etc.
11. January 2014 at 21:59
Erratum factors > factories
11. January 2014 at 23:29
ok suppose the money supply was doubled overnight or in a short period of time. That was be completely imprudent and destructive whether through QE or heli drops.
Going with your example but applied in a efficient manner. You would limit the rate of expansion to something reasonable like maybe 5 or 10% increase in supply per year in regular increments. Under this approach initially demand would increase. Inflation expectations would depend on what the effect of this new demand would be. If there is plenty of slack in labor and productive capacity inflation expectation would only increase marginally. If economy is at potential output and the rate of money expansion is increased new demand would likely result in higher inflation expectations.
If base is increased 5% under heli drops or 5% under OMO’s should result in a similar outcome.
Higher demand will increase the price of financial assets like I explained before.
12. January 2014 at 00:30
lxdr,
I hear what you’re saying, but again it’s not a question of efficiency. As soon as the market were to figure out what is going on, helicopter drops would result in almost instant inflation. It might work the first few times if done in small amounts, but first the fx and commodity traders would figure it out, then producers, then big retailers and unions, and then everyone else. After a few tries, there would be no lag between the base increase and the price increase. POS systems would be hardwired to adjust to changes in the base.
OMO are fundamentally different because the Fed is changing the price of financial assets relative to price of goods and services just in the same way that they would cause apple prices to rise relative to everything else if they went out and bought apples.
If you learned traditional money theory and have been ingrained in QTM, the money illusion, the neutrality of money, etc., it’s very hard to disencumber yourself of the mental model you’ve built and to understand this. And…. it’s doubly hard because there is very good correlation between the quantity of money and NGDP. And… increases in the base nearly always yield some real growth and as well as nominal growth. This near perfect correlation makes the QTM/HPE model seem very good, and it is…. at least from a predictive point of view. But….it is not an accurate description of causality.
IMHO, this is all fairly intuitive if you’re starting from scratch, but for someone with legacy training in monetary economics, it’s a bit like trying to understand bicycle mechanics if you’ve spent many years only ever watching bicycles going downhill and therefore believe that gravity and the motion of the wheels causes the pedals to turn.
I’ve been trying to explain this to Scott for over a year, but he still doesn’t get it…… Hopefully this is an old enough post so that he’s no longer reading the comments. 🙂
12. January 2014 at 17:01
“As soon as the market were to figure out what is going on, helicopter drops would result in almost instant inflation.”
The market is going to figure out that the increase in demand due to heli drops isnt going to pressure prices because of the slack in the economy so prices instantly adjust very little and heli drops will results in small amounts of inflation.
Prices didnt move proportionately before when increasing base under OMO’s so whats going to change if you do the same thing through heli drops? Can you explain this to me I think this is our sticking point.
“OMO are fundamentally different because the Fed is changing the price of financial assets relative to price of goods and services just in the same way that they would cause apple prices to rise relative to everything else if they went out and bought apples.”
Thats becuase they are changing the relative supply demand equilibrium between base and all other financial assets. They increase supply of base and reduce supply of treasuries in OMO’s. Heli drops also change the supply demand equilibrium between assets by increasing supply of money but without needing to interfere in other financial markets.
13. January 2014 at 02:56
lxdr,
I’m going to bow out for awhile after this comment (day job), but maybe we can take it up on a subsequent post. Your points and questions are very interesting ones.
1. First to reiterate, heli drops change the price of everything equally relative to money, but OMO change the price of financial assets relative to goods and services. Also just to be clear, for the purpose of simplicity I think of financial assets as being fungible so that it doesn’t really matter what asset the Fed buys, and when they buy one financial asset (e.g Treasuries), it impacts the price of all financial assets equally.
2. Your question about why price movement would be different under OMO than under a helicopter drop is an astute one. I think the answer is that under OMO there is effectively an obligation to repay (i.e. forgo future consumption). Think about this way. What would be the difference in consumption and prices under two scenarios. In the first scenario, the bank GIVES you 10 cents for every dollar your hold. In the second scenario, the bank LENDS you 10 cents for every dollar you hold.
3. Not to change focus, but there is also another problem with helicopter drops if you believe they don’t cause instant inflation.
[BTW (and just to make sure we’re viewing heli drops in the same way), If you just give money to everyone, I don’t think of that as a helicopter drop. It’s a tax credit or negative tax, which is fiscal policy not monetary policy…. and I think that works pretty much like OMO. The government instead of the private sector borrows more and then spends more.]
I assume that the way you think of a helicopter drop is that the Fed gives money out in proportion to money currently held. In this case if there is not instant inflation it will cause people to horde money in the expectation of future drops and this will have exactly the opposite effect of what is desired.