Pay attention to the theoretical world
Mark Sadowski sent me this link by John Carney:
Sumner, for example, insists on treating quantitative easing as an increase in the monetary base and thinks that must be inflationary. But this is flat-out wrong because it is only looking at one side of the QE ledger””the growth of reserves””while ignoring the other””the shrinking supply of Treasurys.
Again, let’s return to our QE’d investor.
He used to hold a claim on a bunch of securities held by his bank, now he holds a large bank deposit. Sumner and other monetarists insist that he’s more likely to spend this bank deposit, triggering inflation.
But why would he do this? He’s already demonstrated that he doesn’t want to spend the money, that’s why he held the bond to begin with. He wants to preserve his capital for future uses””retirement, bequests to heirs, whatever. The shift from holding a claim on a Treasury to holding a claim on a bank deposit doesn’t change this.
I have two problems here. First, he does not correctly describe my views. I don’t claim that monetary stimulus boosts NGDP and prices because it makes people hold more bank deposits and less bonds. I rely on the hot potato effect for base money. Carney is assuming market monetarists have a Keynesian view of the economy, whereas we actually have a monetarist view. The fact that someone wanted to “preserve their capital” might be important to a Keynesian, but it’s utterly irrelevant to a monetarist. You can preserve your capital without holding lots of currency. It’s not about saving, it’s about hoarding.
More importantly, isn’t Carney claiming that open market purchases in general are not inflationary? That’s a fairly radical claim. I may have misread the post, but that seems to be the implication. In contrast, people usually argue QE is ineffective due to the zero bound problem. Even that is wrong, but for more subtle reasons.
The market clearly reacts to QE as if it is inflationary. I don’t think Carney’s post will cause investors to change their mind.
PS. The reason I argue Carney seems to be making a general argument for the irrelevance of OMOs is that he does assume that QE affects interest rates:
This applies even to highly liquid assets like Treasurys. When the Fed buys securities, it raises the prices of the securities and lowers their yields.
So he’s not relying on a zero bound argument, he’s arguing from a “QE just swaps one asset for another asset” perspective. That’s a radical denial of the effect of any sort of OMO. Or am I missing something?
PPS. Carney also says the following:
When you pay attention to the real world, the situation is much simpler.
Does he think that academic economists don’t understand that open market purchases involve the swapping of reserves for bonds of roughly equal value? Does he understand why we think that matters, even if the bonds were being held for retirement purposes? Yes, pay attention to the real world. But also pay attention to the theoretical world. All real world observation is processed in our brains via theory, AKA “models.” You need the right theories.
Update: Benn Steil sent me the following:
Tags:
11. December 2013 at 13:02
This interview with Gene Fama that CNBC did in late October may be in part what Carney is drawing from:
https://www.creditwritedowns.com/2013/10/economics-noble-winner-fama-qe-is-just-an-asset-swap.html
where Fama said the program didn’t do much to interest rates.
11. December 2013 at 13:49
Hilsenrath says that Obama will nominate Stanley Fischer for Vice Chair. Thoughts?
http://online.wsj.com/news/articles/SB10001424052702304477704579252330777945364?mod=WSJ_hpp_LEFTTopStories
11. December 2013 at 14:34
It seems like the real with John Carney’s post is that he’s taking the “real world” perspective of the average bondholder rather than the marginal bondholder. The marginal bondholder will see the prices of bonds increase due to QE, and determine that they don’t offer as good a risk/reward ratio as other assets. They sell their bonds and invest elsewhere where the risk/reward ratio is more favorable. I think the marginal bondholder would be more likely to move that portion of their portfolio to other assets rather than to keep it in cash. The stereotypical insurance company investing in bonds to match their liabilities isn’t going to the the first $X billions of sellers in QE.
11. December 2013 at 15:26
John Carney:
“When the Fed buys securities, it raises the prices of the securities and lowers their yields…When you pay attention to the real world, the situation is much simpler.”
Evidently paying attention to the “real world” means losing money in the bond market:
http://macromarketmusings.blogspot.com/2013/12/qe-rising-yields-and-right-way-to-taper.html
But at least your situation is “much simpler”:
http://img.gawkerassets.com/img/18j5bmtyxl2khjpg/original.jpg
11. December 2013 at 17:08
Scott,
“It’s not about saving, it’s about hoarding.”
Maybe you’re exaggerating for effect, but these claims seem odd. If “saving” means “not spending,” then “it’s” got to be a bit about “saving.” And if, by “hoarding,” you mean holding some form of asset not produced by labor, then “hoarding” is also part of the problem, but this insight didn’t escape the original “Keynesian.”
11. December 2013 at 17:30
Re: Today’s stock move
So not only is Monetary Policy>>>>>Fiscal Policy, it turns out that the fiscal multiplier is negative!
Keynesians really have some explaining to do.
11. December 2013 at 18:09
I have a different take on Carney’s work, and maybe I am wrong but here goes: Okay, someone sells a bond, and the Fed buys it. That someone now can spend the money, make immediate claims on output of goods and services. They may choose to bank the money but they may also choose to spend it, or re-invest it.
If that someone sells the bond to another person, that other person now has to forego immediate consumption, or re-investing.
But when that someone sells to the Fed, no one has to forego immediate consumption.
It is true bank reserves have swollen under QE, but not quite as much as QE outlays. It is also true people are saving more, and that is swelling bank reserves.
My guess is that some people selling bonds to the Fed end up spending the money. And that is what we want.
11. December 2013 at 18:33
Vivian, Fama says you can do monetary policy simply by controlling the currency stock. I agree, but doubt that Carney does.
Michael. He’s very qualified, did a good job in Israel. Probably a very good choice. That’s all I know.
BJ, I’m not sure it really matters, the bond market isn’t that important in monetary policy. They could do roughly the same thing buying gold or stocks.
Mark, I noticed that too.
Greg, No that’s actually not what the terms mean. Saving is spending money on investment projects, or loaning to people who do. Hoarding is increasing your real holdings of base money. Very different concepts, virtually unrelated. Hoarding often increases during recessions, saving usually declines.
Cameron, I thought it was amusing, but of course I wouldn’t put too much weight on a news headline.
Ben, Maybe, but I don’t think the effect depends on the specific people selling to the Fed ending up spending more, just the fact of QE raising AD.
11. December 2013 at 18:58
Right. Still, one of the reasons the Fed gives for delaying the taper is fiscal uncertainty. This either weakens the case that fiscal uncertainty matters or (more likely IMO) demonstrates how much it pales in comparison to monetary policy.
11. December 2013 at 19:07
Carney’s article seems consistent with Woodford.
11. December 2013 at 19:10
Cameron, Good point.
Bill, Woodford would never say OMOs don’t matter.
11. December 2013 at 21:45
Expansionary monetary policy is effective and is only effective if it causes a marginal increase in the exchange by the non-banking sector of financial assets for goods and services.
Carney argument presupposes that the non-banking sector exchanges assets with the Fed in OMP operations in order to hold more cash. This is not the case. The non-banking sector holds the amount of cash it needs for transactions. A firm or individual does not sell Treasuries (through intermediaries) to the Fed simply to hold more cash, anymore than a firm or individual takes out a bank loan in order to hold more cash. Firms and individuals sell treasuries, take out bank loans, etc. in order to buy goods and services. If the non-banking sector is not induced (by higher asset prices or higher NGDP expectations) to increase their exchange of financial assets for goods and services then there will be no increase in money held by the non-banking sector.
It is not the holding of additional money which induces the non-banking sector to increase nominal spending. Rather the intention to increase nominal spending is what causes the non-banking to hold higher money balances. Unless there is an increase in intended nominal spending, money exchanged for assets in QE/OMP will simply end up as higher ER balances by the banking sector.
11. December 2013 at 21:50
The sad thing is that I would agree with the idea “QE isn’t inflationary or deflationary”. So yes, if you pay attention to the real world QE was inflationary. Now for why it was so one has to have some grasp on the theory.
BTW this strongly reminds me all those comments like “Hahahahaha academic economists think that banks lend reserves. They do not know that it is not how monetary policy operate in the real world [insert any wacky theory here]”
12. December 2013 at 01:15
I think Woodford (with Curdia) argued that targeting certain segments (lowering spreads) matters more because “ordinary” OMOs are subject to Wallace neutrality. Call it “credit easing beats quantitative easing”. Did I understand that right?
12. December 2013 at 04:36
Scott:
I don’t think Carney was saying that Open Market Operations don’t matter. Their impact on base money doesn’t matter. Only their impact on interest rates. And long rates are the future short rates plus a liquidity premium. And so all Quantitative Easing adds beyond the future course of short rates is a reduction in the liquidity premium. This could have some effect, but it is doubtful. Woodford does say that, and as best I can tell that is based on Wallace Neutrality (nonsense.) The evidence that we Market Monetarists point to is also used by Woodford to show that they aren’t effective. They aren’t in fact lowering interest rates. And only by lowering interest rates would they possibly raise current aggregate demand.
We frame this as they are effective even though they don’t lower interest rates. The effect on future nominal GDP raises the equilibrium real and nominal interest rates. Well, the actual increases certainly cast doubt upon QE lowering interest rates, and lower interest rates raise current consumption.
I don’t agree with Woodford’s framing, but I more and more agree that Quantitative Easing is not very effective in theory. There is more than enough base money to meet the Fed’s current target. Excess money that will be withdrawn if it has any effect won’t have any effect. Or not much of one.
Suppose we had a nominal GDP level target, and nominal GDP was on target. The Fed decides to do quantitative easing to lower the unemployment rate. The Fed _is_ committed to the target. It is doing the Quantitative Easing conditional on nominal GDP not rising above target. If it does, they will reverse course and bring nominal GDP back down to the target level. And, people believe this. We have a tight, Market Monetarist target.
Now, I don’t think Quantitative Easing would lower the unemployment rate. But I also don’t think it would push nominal GDP above target much, if at all. The quantity theory exercise of what happens if the quantity of base money is the nominal anchor and it permanently increases just doesn’t apply.
Now, flexible inflation targeting is much looser than nominal GDP level targeting. And so, it is more likely that Quantitative Easing could have some impact on spending on output, and so, inflation and real output. It might make it slightly more likely that the price level will shift up to a higher growth path. And the Fed won’t reverse that because it is doing inflation targeting.
This theory suggests that it is the Evan’s rule, saying that inflation will be allowed to rise temporarily, that has had the biggest effect. If it actually happens, it is a shift up in the growth path of the price level.
Naturally, I have been thinking our preferred target and how it would work if there were “excessive” increases in base money. It seems to me that there is a lot of leeway.
12. December 2013 at 05:34
Scott,
Have you ever commented on David Einhorn’s “Jelly Donut” criticism of QE? He wrote this in May 2012 so it has been around for a while. I would like to highlight two points he makes that point out how forcing interest rates lower actually harms the real economy.
#1: QE destroys the time value of money which promotes economic procrastination, rather than urgency. “Why should anyone make a marginal decision to borrow and spend or build today, knowing that low-cost financing will still be available through the end of 2014″?”
#2: QE boosts the balance sheets of corporations, by lowering their interest expense, while destroying the interest earnings of individuals. This process destroys aggregate demand as individuals have LESS income to spend than they otherwise would in a normal rate environment.
http://www.huffingtonpost.com/david-einhorn/fed-interest-rates_b_1472509.html
12. December 2013 at 06:20
ps.
In truth changing interest rates has no effect on total interest income. However, changing interest rates does change the composition of who receives that income. Corporations are poised to spend extra income anywhere in the world. Individuals are more inclined to spend extra income locally, such as at a local retailer or dealership.
One of the many blindspots in macroeconomics is the assumption that there is such a thing as aggregate demand. AD is a theoretical concept that does not actually exist. Every individual has a unique demand curve and the preferences of the individual are vastly different than the preferences of corporations. As such when making economic predictions (which is what economic modeling is all about) it matters a lot to differentiate between who is earning what. Lowering money market rates from 3% to 0.5% may do wonders from Goldman Sachs and its partners but for Joe & Jane it pinches their finances. And in the real world there are a whole lot more Joes & Janes than there are partners at Goldman Sachs.
12. December 2013 at 06:23
Dan W.,
“This process destroys aggregate demand as individuals have LESS income to spend than they otherwise would in a normal rate environment.”
Great: we can raise interest rates to 50% and get all the stimulus we need. We’d better avoid raising them too much higher, though; we don’t want hyperinflation.
12. December 2013 at 06:27
And why didn’t policymakers in the 1970s see that they could get inflation under control by cutting interest rates? (Not too much, though: cutting them to 0.5% in 1975 could have caused a depression.)
12. December 2013 at 06:50
Bill, Carney clearly says QE does not raise inflation, so I’m certain you are misreading him. And he also says QE reduces interest rates. That’s not at all what Woodford believes.
As far as the rest of your comments I certainly agree that temporary QE is not inflationary. But the Fed would be insane to do a QE program that is 100% temporary, and the markets clearly understand that it is partly permanent.
Dan, Totally disagree. He thinks low rates are due to easy money. All we need to do is look at the eurozone and Japan to know that is completely wrong.
You said;
One of the many blindspots in macroeconomics is the assumption that there is such a thing as aggregate demand.”
I define AD as NGDP, which certainly does exist.
12. December 2013 at 06:58
W.Peden,
Is there a difference between Ben Bernanke deciding what the price of money should be and FDR deciding what the price of gold should be? Both are exercises in hubris if you ask me.
That said if higher interest rates are so destructive to economic growth how did our economy perform so well from 1994 – 2000 despite an increase of the federal funds rate from 3% to over 6%?
The evidence continues to grow that zero interest policy is a a self-imposed misery. Volker jacked interest rates up in the early 1980s to force a reset of a broken economy and this paved the way for nearly two decades of growth. We need a similar reset but crony capitalism and bad economic theory is preventing it. Rather QE infinity is a corporate handout that artificially increases assets prices, destroys return on assets and in fact is a deflationary influence.
And the shame of it all is that Japan has shown us how zero interest rate policy plays out. Why would we assume our outcome would be different?
12. December 2013 at 07:14
Dan W.,
“Is there a difference between Ben Bernanke deciding what the price of money should be and FDR deciding what the price of gold should be? Both are exercises in hubris if you ask me.”
Deciding what the price of money should be and targeting a price index are one and the same thing.
“That said if higher interest rates are so destructive to economic growth how did our economy perform so well from 1994 – 2000 despite an increase of the federal funds rate from 3% to over 6%?”
Come on! This is basic monetary theory.
“Volker jacked interest rates up in the early 1980s to force a reset of a broken economy and this paved the way for nearly two decades of growth.”
I don’t even think it’s worth beginning.
“And the shame of it all is that Japan has shown us how zero interest rate policy plays out. Why would we assume our outcome would be different?”
The causality in Japan entirely the other way around: an absence of inflation causes low interest rates.
Two words summarise my thoughts reading these arguments- “Fisher effect”. Unless someone is willing to bring that up and discuss it, or at least demonstrate that they’ve heard of it, I’m not even sure it’s worth proceeding.
12. December 2013 at 07:21
Dan W wrote:
“Is there a difference between Ben Bernanke deciding what the price of money should be and FDR deciding what the price of gold should be? Both are exercises in hubris if you ask me.”
Interest isn’t “the price of money”. The price of money is whatever someone is willing to give you in exchange for money.
“That said if higher interest rates are so destructive to economic growth how did our economy perform so well from 1994 – 2000 despite an increase of the federal funds rate from 3% to over 6%?”
Are you suggesting that 1994-2000 and 2008-2013 were identical, but for the difference in interest rates?
“The evidence continues to grow that zero interest policy is a a self-imposed misery. Volker jacked interest rates up in the early 1980s to force a reset of a broken economy and this paved the way for nearly two decades of growth.”
If you can’t see some obvious differences between the problems affecting the late 70s economy and those affecting the economy of the past five years, you really need to look harder.
“Rather QE infinity is a corporate handout that artificially increases assets prices, destroys return on assets and in fact is a deflationary influence.”
This makes no sense whatsoever. The smart thing to do with any asset whose price is “artificially increased” while its return is “destroyed” is to sell it!
Do you think the 70s Fed screwed up by not engaging in massive QE to get prices down?
12. December 2013 at 08:00
Michael Byrnes,
How can one argue with someone who takes reductio ad absurdums as premises? It’s like trying to freeze ice.
12. December 2013 at 08:16
I’ve tried to freeze ice many a time.
12. December 2013 at 08:22
One might say that an interest rate is the nominal price of doing nothing with money. A zero interest rate means the nominal price of doing nothing is very low. Einhorn explains this as one of the paradoxes of zero interest policy. A higher interest rate would actually encourage the holders of real assets to do develop them or sell them to someone who would. Of course selling would likely lower the price of that asset and in fact prices would lower until the expected rate of return on buying the asset met the market’s expectation.
When interest rates are zero there is little incentive for banks to markdown foreclosed properties. So these assets sit dormant and as such DO NOT CREATE VALUE!!!!!!!!! And as a result economic growth stagnates.
While it is true that a nominal zero interest rate policy can create a real negative interest rate this condition does not require a zero interest rate! Anytime the funds rate trails the CPI the real interest rate is negative.
Low, and especially negative, real interest rates are indicative of poor economic growth. This is a situation one should want to get out of! So how does it make sense to encourage a condition that one wishes to avoid?
Appreciate that what Volker did in 1981 was to force real interest rates positive in a HUGE way. Savers in the early 1980s were hugely rewarded. Their incomes increased. They felt richer. On the other hand equity and bond valuations hit bottom, but from there the opportunity for long-term economic growth was created.
Forcing real positive interest rates worked in 1981. Perhaps this is the remedy to economic malaise. If so why not give it a try now?
12. December 2013 at 10:59
Scott,
Thanks for taking notice of my post about QE. (And thanks to Mark for bringing it to your attention.) I’m going to try to address your two points here.
1. Hot Potato Effect. First, I’m sorry if I mischaracterized your argument. I don’t think your Hot Potato effect works, however.
You’ve used the example of new gold discoveries to explain HPE. http://www.themoneyillusion.com/?p=23314
But a better analogy to contemporary monetary economics would be bimetallism.
Let’s say that an ounce of gold is priced at some amount silver ounces. Silver is the general medium of exchange but a lot of folks have their savings in gold because storing that much silver is cumbersome.
The central bank announces that it will start buying gold in exchange for silver as part of its Commodity Easing program. The supply of gold in public hands is reduced, the supply of silver is increased. This should alter the relative prices of gold and silver by a bit but not by very much because, after all, they’re close substitutes. Silver is more liquid and useful for buying stuff, but gold is less liquid but also costly to store. You gain in liquidity what you lose in ease of storage. Market inefficiencies, in other words, explain the portfolio preferences for one or the other.
What happens to the overall price level in everything else? Nothing very much. The total supply of money””gold and silver””is basically the same. The composition of the money supply has changed. But no one has more of the money metals than they wanted to have before.
This is what is happening when the Fed engages in QE. The composition of the public’s holdings change to more reserves, less Treasuries. But the overall supply of safe assets remains in equilibrium. No one has more of the money papers than they wanted before.
To put it slightly differently, your HPE breaks down because it doesn’t recognize the money-ness of Treasuries. The problem is that you are looking at traditional measures of money and mistaking those maps for the territory.
2. Open Market Operations. Bill Woolsey is right about my position. It’s not that OMOs don’t matter, it’s that they matter because of their effect on interest rates rather than on the size of the money supply.
By the way, Bill, I’m open to the idea that the effect on term premia is exaggerated or non-existent. But for now I think I’m persuaded to the side of Ben Bernanke on this.
From Bernanke’s November 19th speech:
http://www.federalreserve.gov/newsevents/speech/bernanke20131119a.htm
“LSAPs, in contrast, most directly affect term premiums. As the Federal Reserve buys a larger share of the outstanding stock of longer-term securities, the quantity of these securities available for private-sector portfolios declines. As the securities purchased by the Fed become scarcer, they should become more valuable. Consequently, their yields should fall as investors demand a smaller term premium for holding them. This argument depends importantly on the assumption that the longer-term Treasury and MBS securities that the Fed buys are not perfectly substitutable with other types of assets, an assumption that seems well supported in practice.”
12. December 2013 at 11:03
Carney,
Money which doesn’t have price risk is crucially different than USTs that always have some price risk. The HPE on money is a rising price level. But the HPE effect on USTs is rising interest rates. Completely different phenomenon.
12. December 2013 at 11:14
Ed–Treasuries have price risk. Ask a bond trader.
12. December 2013 at 11:14
Oh, sorry, Ed. Misread your post. Of course money has price risk, that’s where you get the HPE from.
12. December 2013 at 11:16
Carney,
Re-read my comment. That is what I said. Unlike money, USTs always have some price.
I am a bond tradeer.
12. December 2013 at 11:19
Dan W wrote:
“Forcing real positive interest rates worked in 1981. Perhaps this is the remedy to economic malaise. If so why not give it a try now?”
Because it isn’t 1981? The economy of 2009 was nothing like that of the late 70s. Nothing like it. The economy of 2009 was more like the economy of 1929 – tight money was tried in 1929 to disastrous effect.
Europe triggered a double dip by raising rates in 2011. No thanks.
12. December 2013 at 11:19
Carney,
How could you lose money (nominally) holding US Dollars?
12. December 2013 at 11:25
“Forcing real positive interest rates worked in 1981. Perhaps this is the remedy to economic malaise. If so why not give it a try now?”
It did a good job of causing a severe recession in 1982, along with a drastic disinflation. If that’s really what we wanted now, it would be easy to get.
12. December 2013 at 12:17
John Carney wrote:
“Of course money has price risk, that’s where you get the HPE from.”
I think the differences is that a dollar bill will always be worth $1.00 in nominal terms. The purchasing power of that dollar will change over time, but there will *never* be a time when you cannot exchange it for $1.00, or for $1.00 worth of goods and services.
That’s different from a bond; a bond’s price (in dollars) will not remain constant over time. If you buy a bond for $1,000, you have no assurances that you will be able to sell for $1,000.
12. December 2013 at 13:15
Ed: I agree with you that dollars won’t lose value in nominal terms. As I said, I misread your earlier comment.
Michael: Sure. So what?
12. December 2013 at 14:40
John C — interest rates are mostly the product of (monetary base / NGDP) (log-log). This framework — which applies all the way back from 1918 to now — means that sub-1% rates on the short end always requires huge production in the monetary base.
Whether this is a supply- or demand- thing, we don’t know. More to the point, you don’t know.
Are banks demanding reserves, or households demanding cash, or is the Fed is forcing down interest rates by purchasing bonds? You don’t know, and cannot know.
Note that rates successfully went quite low in the 1930s when revalued gold was the major holding, and Treasuries were less of a factor on the balance sheet.
I’d suggest that interest rates are a price outcome in the largest market on earth; and that we are all — including the Fed — price takers. This means that there is barely any interest rate effect; and all that matters is the monetary base/NGDP. You can trade on speculative expectations all you like, but you will get killed if you get base/NGDP wrong.
BTW, this safe asset thing is a bit of a tautological dead end. (Imagine gold at $1,000,000/oz — that would be an immense safe asset pool with no counterparty risk. Bottom line, there are unlimited and infinite safe assets.)
But, if you mean safe assets in a narrow conventional fixed income collateral sense, consider: more leverage means fewer safe balance sheets (lower credit quality). If an economy is overleveraged, its no surprise that it is difficult to find collateral to add leverage further.
It’s not a shortage of safe assets, it’s an abundance of preexisting leverage that matters.
12. December 2013 at 15:05
Paden, Michael & Scott,
What should be the consequence on the national economy when fraud, bad debt and bad decisions on the order of trillions of dollars come to light? Is the fallout from such error something that can be covered over by monetary easing? If it can then why should we ever worry about bad debts or financial fraud? Heck, why work if debt can be increased without end?
Of course you know work is a necessity and that there must be a consequence to bad debt. Experience says that the consequence is deflation. Assets that were purchased with money that cannot be repaid will be reclaimed and resold and usually at a fraction of the previous price. Economists used to call this the business cycle.
Does NGDPLT allow for the business cycle? If so how does one decide what part of the business cycle represents the level target?
How is “extraordinary monetary policy” that prevents the repricing of assets to lower levels not fraud? This is a serious question. If the huge appreciation in house prices or dotcom stocks is directly related to fraudulent business activity then how is it appropriate to rely on monetary intervention to prevent these assets from selling at lower prices? And how is it appropriate to intervene to prevent related industries, that profited indirectly from the fraud, from suffering financial loss?
12. December 2013 at 16:38
Dan W.:
Have you read Scott’s article, “Re-Targeting the Fed”? It addresses a lot of your questions.
http://www.nationalaffairs.com/publications/detail/re-targeting-the-fed
12. December 2013 at 18:21
Carney’s argument can be rather easily dismissed by realizing that just because investors in bonds dispovelay a penchant for saving and investing rather than consuming, it doesn’t mean that “swapping” bonds for reserves isn’t going to affect prices all that much.
For we have to consider the full implications of government borrowing. In order for the Fed to buy bonds from the banks, the banks have to have already displayed a penchant for NOT hoarding money, but rather a penchant for investment spending. That money is collected by government and spent into the economy, raising spending and prices.
It doesn’t matter if the banks then hoard cash sent their way by the Fed as the Fed buys those bonds. There has already been spending by virtue of the expectation that the Fed will buy more of the bonds the banks buy.
The government buys money from those who don’t want to spend that money on consumption. The government then spends that money. All this is predicated on the banks being able to sell bonds to the Fed later on.
12. December 2013 at 19:44
Dan, You asked:
Is the fallout from such error something that can be covered over by monetary easing?”
No.
John Carney, You said:
1. QE lowers interest rates.
2. Monetary policy is only effective if it lowers interest rates.
3. QE is not effective.
What am I missing? I must have misinterpreted you on one of the three points.
BTW, I do not believe monetary policy is only effective if it lowers interest rates.
And I do not consider money and bonds to be close substitutes. When I go shopping I do not agonize about whether to bring cash or T-bills to the grocery store. And bonds pay interest, cash does not. Even reserves paid no interest until 2008, when the Fed shot itself in the foot with IOR.
And don’t worry about misinterpreting me, almost everyone does. I’m used to it as I have a unconventional view of things.
13. December 2013 at 01:57
Dan W.,
“Is the fallout from such error something that can be covered over by monetary easing?”
As Scott says, no, and providing general liquidity to keep NGDP on track is distinct from bailing out one particular sector.
13. December 2013 at 11:57
Scott,
Thanks.
1. I see evidence that QE lowers rates but the mechanism is still unclear to me. Is it portfolio effects and term premia (as Bernanke says)? Communication influencing long-term rates? Reflexivity (people think it will lower rates, so it does)?
This doesn’t tell us by how much QE lowers rates.
2. Monetary policy can be “effective” outside of rates rising or falling. I think the main thing OMO do is push direction of rates through signaling, not quantity.
3. Whether QE is effective depends on what we mean by effective. It certain works to lower quantity of Treasuries, raise reserves, raise deposits. Does that have broader effects on financial markets? It appears to, although again the mechanism is unclear. Portfolio effect? Reflexivity? Does it have broader effects on economy? Perhaps. Mechanism unclear, again.
Reflexivity may be doing a lot more work here than people realize.
13. December 2013 at 14:59
John Carney, reflexive prices change everything, yes, but wouldn’t that also change the price of any free lunches that the Fed might hand out?
What would your reflexive analysis be if the Fed were indeed handing out free profits to bond sellers? Should bondholder sell, then? What would happen to yields?
Consider also that bonds are also being judged versus other asset classes — so the quantity of treasury bonds doesn’t signify.
Bottom line, QE lowering rates is pure bunko. I go so far as to say that the Fed buys bonds with the express intent (deep and liquid market) of not distorting prices when they buy or sell. The liability creation side — currency and reserves — is the Fed’s power, not its asset side.
Rates are really only the price of debt. Too much leverage, too few safe assets, means that there are few- to no- creditworthy borrowers — hence the price of existing debt by the few creditworthy borrowers rises quite high (low yields), and banks begin to substitute less-risky reserves instead of price-risky-and-rich notes.
14. December 2013 at 07:42
John, OK, but I’m still kind of confused as to why you think it does not raise inflation. You see it affect markets in ways consistent with it being expansionary. You are not sure why. Fair enough. But that’s no reason to deny it’s expansionary effect. Monetary policy always works through very subtle channels, even when not at the zero bound.
18. December 2013 at 20:07
John Carney & ssumner, let’s try it this way.
IORR and IOER are both zero. Fed funds rate is zero. Ten-year treasury yields 2.75%. Fed announces QE for the ten-year at 2.75%. The seller is an individual (outside the banking system) for $1 million. All banks have the central bank reserves they want and enough goods/services to run. The fed keeps the reserve requirement where it is, and no new loans are made. The bank of the seller gets its account at the fed marked up by $1 million in central bank reserves, and the seller’s checking account gets marked up $1 million in demand deposits. The seller does nothing with the checking account.
Exactly how do the $1 million in central bank reserves move?