Is Robert Hall a Sumnerian?
A recent post by Arnold Kling contained this passage from an interview of Robert Hall:
even though the Fed has driven the interest rate that it controls to zero, it hasn’t had that much effect on reducing borrowing costs to individuals and businesses. The result is it hasn’t transmitted the stimulus to where stimulus is needed, namely, private spending.
Immediately after the quotation, Arnold had this to say:
The spread between the interest rates on private debt and the interest rates on Treasuries shot up during the crisis. Hall calls this an increase in “friction.” Many of us instead call it an increase in the risk premium. Whatever one calls it, it is an important phenomenon. No matter how hard he tries, Scott Sumner just cannot convince me that this phenomenon is due to a drop in expectations for nominal GDP growth.
Perhaps I am reading too much into this juxtaposition of comments, but it seems to me that Kling is implying that Hall’s views of the crisis are different from mine. Just to be clear, I think that credit markets did freeze up in late 2008, and that that hurt the real economy. But I also think the Fed could have greatly reduced the severity of the financial crisis. And I am pretty sure that Hall agrees with me. Here is Robert Hall and Susan Woodward:
Raising the reserve interest rate is a contractionary measure. A higher interest rate on reserves makes banks more likely to hold reserves rather than increasing lending. The Fed’s decision to raise the reserve rate from zero to 75 basis points just as the economy entered a sharp contraction in activity is utterly inexplicable. Fortunately, the Fed lowered the reserve rate subsequently, but the continuation of a positive reserve rate in today’s economy is equally inexplicable. Some economists have proposed that the Fed charge banks for holding reserves, an expansionary policy worth considering. With the Fed funds rate at around 15 basis points, it would take a charge to restore the differential that drives banks to lend rather than hold reserves. Were the Fed to charge for reserves, they would become the hot potatoes that they were in the past, when the reserve rate was zero and the Fed funds rate 4 or 5 percent. Banks would expand lending to try not to hold the hot potatoes and the economy would expand. There is no basis for the claim that the Fed has lost its ability to steer the economy. (However, the Fed would have to go to Congress to get this power, as it did to get the power to pay positive interest on reserves.)
Isn’t that precisely my view of the crisis?
1. Interest on reserves was a big mistake and still is.
2. If anything the rate should have been negative.
3. The Fed did not lose it’s ability to steer the economy when rates hit zero.
I think the third point implies that if we aren’t happy with the ditch into which the economy has been “steered,” we should take a hard look at what the Fed has or has not been doing.
Tags:
4. July 2010 at 17:05
I’ve wondered if perhaps it wouldn’t be possible to more simply charge a negative rate on excess reserves and a high positive rate on required reserves to deter cash liquidation and encourage lending.
Its not obvious to me what this policy would do, but it seems to me that since required reserves are illiquid per se, that the market-rate of interest would decline to respect the excess reserve rate not the required reserve-rate. The required reserves being mandatory anyways holding them is not in lieu of other investments.
And in the short-run, banks would rollback their sweep programs and immediately convert everything they could into something having a reserve requirement.
Suddenly then the Fed would be pumping a very large stream of interest payments into the banking system, e.g., in the billions which would certainly be a big boost to bank profits, and therefore capital.
4. July 2010 at 17:44
Jon, I made that proposal last year. I made the proposal in response to criticism that my negative IOR plan would hurt bank profits during the crisis. But you are right, it would encourage the creation of bank deposits, and thus raise the money multiplier.
4. July 2010 at 17:48
Playing Devil’s Advocate – what if the someone from the Fed said, “Just between you and me, the real problem is that the banking system still is much more fragile than we feel is prudent to admit in public. This admission, we feel, would cause all kinds of panic with real detrimental effects of uncertain magnitude.
The large amount of excess reserves and the interest rate we pay banks to hold it is, unfortunately, indispensable to keep the banking system afloat for the at least the next 12-18 months until the large institutions have enough opportunity to recapitalize and digest their massive portfolios of non-performing loans.
We understand what this is doing to the economy, deflation, and unemployment, and we feel the country’s pain. However, in our judgment, the pain could easily be *much worse* if we were to goose banks to do things they insist are imprudent and high-risk – and then one of them goes belly-up starting a whole new cascade of potential failure and disarray. And this time – people won’t believe we can save the system because, well, nothing would seem to prove them more correct than another big system failure.
So, like we said – big banks are still in trouble and this is designed to help them, at enormous cost, but a cost we feel is worth bearing, because failure would be worse.”
Does this sound at all plausible? What’s the best counterargument?
4. July 2010 at 19:10
Excellent post. I read the Hall interview.
Long ago, the Reaganites wanted to edge Volcker out (a Carter appointee, originally), and run roughshod over David Stockman, OMB chief and former theology student. They did both, and we got both fiscal and monetary stimulus, at a time of much higher inflation that we have now. George Gilder (Wealth and Povery) was giddy at this. In some ways, it worked, and the economy grew, and inflation came down (we did build up debt for taxpayers).
According to Hall, we are not fiscally stimulating now, as local and state governments are cutting back, offsetting the federal government’s outlays.
Seems to me the Fed needs to blow the doors open. Inflation is dead, dead, dead, deader than Jimmy Hoffa dead. We have much lower expectations that the Reagan days.
Time for some real hard-hitting stimulus. BTW, I read that China’s money supply up 22 percent annually. And their economy is growing.
4. July 2010 at 19:41
‘The Fed’s decision to raise the reserve rate from zero to 75 basis points just as the economy entered a sharp contraction in activity is utterly inexplicable. Fortunately, the Fed lowered the reserve rate subsequently, but the continuation of a positive reserve rate in today’s economy is equally inexplicable.’
Not inexplicable at all
Here are some possibilities:
1.) The fed was trying to constrain ppi inflation that was skyrocketing. They attributed it to demand side, instead of realizing it was mostly an adverse supply shock (possibly caused by corn ethanol laws).
2.) The fed isn’t actually there to help the economy but to protect the banks.
I’m only a grad student and I could even come up with that.
5. July 2010 at 08:41
Hall and Woodward have no idea how the monetary system works.
Banks don’t “lend reserves”.
Check any Post Keynesian website for the explanation.
5. July 2010 at 09:36
anon:
you are wasting your time with sumner. he has no clue how the monetary system works either, and if he ever figures it out, he will need to flush his entire life’s output to date down the toilet.
he is not that strong.
5. July 2010 at 09:56
Professor Sumner,
I’m sure this is an easy one for you, but I’ve been meaning to ask you the following…
If money was tight, according to a basic money market supply and demand model, this means that demand for money went up. What exactly caused money demand to go up?
Best,
Joe
5. July 2010 at 10:35
Indy, There are good counterarguments for all of those points. The negative interest on excess reserves can be accompanied by positive interest on required reserves—hence there need not be any hit to bank profits. And the action would not encourage banks to make bad loans, they could simply exchange ERs for T-securities.
Benjamin, You are even more pro-stimulus than me. But I am confused about the history. I have read in many places that Reagan supported Volcker’s anti-inflation policies, and indeed re-appointed him.
Doc Merlin, PPI inflation was plunging in late 2008 and 2009.
anon and zanon, Hall knows more monetary economics than all the post-Keynesians of the world combined. If I go to a bank, get a $10,000 loan, and withdraw it in cash, then the bank has “lent reserves.” True, the process is usually more indirect in practice, but in the long run the effect is the same. Hall knows what he is talking about.
Joe, During a financial crisis there is increased demand for liquid assets like money. The same process occurred in the Great Depression.
5. July 2010 at 13:40
One year ago Hall was 100% Sumnerian. But now he is much more creditist than you are.
5. July 2010 at 14:26
@Scott
“Doc Merlin, PPI inflation was plunging in late 2008 and 2009.”
PPI was skyrocketing in early 2008/late 2007, and at least according to FRED that is when the recession started.
5. July 2010 at 14:29
Scoot-
Volcker was first appointed by Carter. The Reaganites never really liked Volcker, and replaced him with Greenspan in 1987. The Reaganites were hoping for a money and fiscal surge to carry Bush Sr, into office, and, who knows, maybe it worked (Bush Sr. did win, of course).
Others say Volcker was fired for having independent views on bank regulation. “Paul Volcker, the previous Fed Chairman known for keeping inflation under control, was fired because the Reagan administration didn’t believe he was an adequate de-regulator.” Stiglitz (Wikipedia).
The record is that when Reagan was in office, the Reaganites wanted both fiscal and monetary stimulus. If you go back and read “Wealth and Poverty” by George Gilder, in it you will find Gilder rhapsodizing about inflationary booms. “Deficits don’t matter” became an R-party battle cry.
I bring this up to highlight the role that politics plays in policy. pronouncements, not to suggest that Democrats are any better than Repubs.
The “right-wing: is bashing Obama for deficits and calling for right money. You would hear a different tune–and did hear a different tune–when the R-Party was in power.
I completely agree with you that the tim is past due to try unconventional monetary stimulus. This may be a recession for some, but it is a depression in property markets and for anyone looking for work.
5. July 2010 at 16:38
“If I go to a bank, get a $10,000 loan, and withdraw it in cash, then the bank has “lent reserves.””
Gee, I thought the issue was banks hoarding reserve BALANCES, since that’s what they receive interest on. They don’t receive any interest on their vault cash. Your example demonstrates Anon’s and Zanon’s point, actually–the bank didn’t need any reserve BALANCES to make the loan and create the deposit. How does payment of interest on reserve balances in any way deter the bank from making this loan to you?
5. July 2010 at 18:23
123, Where is the evidence that he has changed his mind?
Doc Merlin, We weren’t talking about the beginning of the recession, but rather the interest on reserve program, which was in effect during late 2008 and 2009, when the PPI was falling.
Benjamin, I have a lot of problems with your account.
1. You never disputed my assertion that Reagan supported Volcker’s anti-inflation policy.
2. Reagan reappointed Volcker in 1983, something he certainly wouldn’t have done if he was worried that tight money was endangering his re-election (recall unemployment was very high at the time Volcker was re-appointed.)
3. Most supply-siders favored stable prices, not high inflation.
4. Gilder didn’t make decisions for the Reagan administration.
So I am afraid I can’t agree with your account. It was Carter that put pressure on Volcker in 1980, which is why he backed off and we had to have two recessions, not one, to get inflation down to 4%. Reagan supported his tight money policy.
But at least we agree that we need easier money now.
Turd, I appreciate your comment. I always enjoy responding when someone adopts a sarcastic tone, and then says something silly.
As I am sure you know Turd, banks prefer to hold a given stock of vault cash, depending on their customer needs. When someone withdraws $10,000 cash, the bank will generally ask the Fed for another $10,000, pulled out of deposits at the Fed. And this extra cash is used by the bank to restore its vault cash to the desired level. So it is effectively a loan of reserve balances, which gets converted into cash at the borrowers request. It’s all base money, all interchangeable. Banks and the public, not the Fed, determine the use of base money (at any given IOR.)
And the IOR is important because it is the opportunity cost of lending out excess reserves. If a bank earns 2% interest on its excess reserves, it is not very likely to lend it to the government (i.e buy T-bills) at a rate of 1/2%.
5. July 2010 at 19:16
Way to change the subject. I didn’t ask why the bank wouldn’t buy a Treasury. I asked why the bank wouldn’t make the loan to you given the payment of interest on reserve balances.
In the end of your example, the bank ends up with the same amount of vault cash after converting its reserve balances. There’s no net change to the asset side (+10k in loans, -10K in reserve balances, vc same as before) and no net change to the liability side (deposits up 10k and then down 10k as you withdrew the deposit).
So, as you said, the bank just converted its reserve balances into a loan. Again, how is it that the payment of interest on reserve balances was a disincentive from doing this? They were earning 0.25% on the reserve balances, and will now earn the prime rate or whatever on the loan. If the borrower is creditworthy, most banks should want to make the loan.
But let’s go more directly to Anon and Zanon’s point. Let’s assume that the bank didn’t have the 10K in reserve balances to convert into vault cash. Then what? Is the bank somehow unable to make the loan and then cover your withdrawal? No. The bank buys the vault cash from the Fed and receives an overdraft to its reserve account, which it clears by the end of the day, probably through the fed funds market or other money markets. Then the final tally is an increase in loans of 10K on the asset side and an increase in money market borrowings of 10K on the liability side. The bank didn’t need the reserve balances to make the loan, and in fact only had an overdraft in its reserve account that it subsequently cleared. And, in fact, this is how the lending process generally works in the US.
And that was their point. It doesn’t matter if banks “hoard” reserve balances because they don’t need them to make loans and create deposits in the first place. There’s nothing about paying interest on reserve balances that will keep them from making a loan to a credit worthy borrower, since they don’t need to have the reserve balances in the first place to make the loan or clear the withdrawal of a deposit.
5. July 2010 at 19:49
“Doc Merlin, We weren’t talking about the beginning of the recession, but rather the interest on reserve program, which was in effect during late 2008 and 2009, when the PPI was falling.”
Let me explain further:
They announced they would pay interest on reserves October 6, 2008. The act (Emergency Economic Stabilization Act of 2008) that allowed them to do it (they had been pushing for this for a while) was passed on October 1, 2008. Looking at the minutes from October 28-29th (the minutes from the 3 earlier meetings are not available) it seems like they are explicitly trying to keep interest rates higher by establishing a floor for them. In fact they seemed dismayed that it didn’t work as well as they wanted it to.
‘While the payment of inter- est on reserves seemed to be helpful in mitigating downward pressure on the funds rate, a number of in- stitutions evidently were willing to sell funds at interest rates below that paid on excess reserve balances.’
http://www.federalreserve.gov/monetarypolicy/files/fomcminutes20081029.pdf
Their projections they give in the minuted totally miss the possibility of price deflation (even as it had been happening for a couple months) and predict price inflation for 2008. In short, it looks like the federal reserve is /very/ backward looking, and didn’t see the deflation happening as it happened. This is more rank incompetence than actual malice, IMO.
5. July 2010 at 20:18
@turd ferguson
Wrt your example:
Interest rates on reserves is a policy designed to put a floor price on interbank lending. This raises borrowing costs for the bank and encourages a bank to keep money in reserves with the fed.
This then raises the prime rate beyond what it would be if the interbank lending rate was below what the fed set as a floor, and causes MB to rise without strongly affecting higher aggregates. Remember banking is a competitive industry, so the prime rate is very close to actual costs (including risk and opportunity costs) for the bank.
Anyway here is evidence of what Scott was saying about paying interest on reserves being contractionary.
http://picasaweb.google.com/lh/photo/0qCIGAFGywrCyxAQjcBIDnADnhk_L4Rt11PaKllIWcg?feat=directlink
As to your point:
“And that was their point. It doesn’t matter if banks “hoard” reserve balances because they don’t need them to make loans and create deposits in the first place. There’s nothing about paying interest on reserve balances that will keep them from making a loan to a credit worthy borrower, since they don’t need to have the reserve balances in the first place to make the loan or clear the withdrawal of a deposit.”
1. If a bank has to borrow money to make the loan instead of loaning it out from reserves it increases costs to itself. What paying interest on reserves does is increase the opportunity cost of using reserves.
2. Banks have a reserve requirement. They simply aren’t allowed to loan out more than a certain multiple of their reserves. Nor are they allowed to borrow more than a certain amount either (this is why the bailouts where in the form of preferred stock that paid 6% interest, instead of a loan).
These two facts combine to have an effect that means that banks may as well be loaning out reserves. Are they in reality loaning out reserves? Not according to the accounting, but they may as well be. This is why monetary economists treat it as if they are.
5. July 2010 at 20:38
Doc,
You’ve completely missed the point. The rate paid on reserve balances is equal to the target rate. Bank borrowing costs are based on the target rate. There is no change to costs as a result of interest paid at the target. And the chart is not evidence of anything except the fact that the Fed created a bunch of reserves through its standing facilities. The chart is perfectly consistent with my explanation above. Your points 1 and 2 are also incorrect.
5. July 2010 at 20:39
Turd:
Please do not waste scotts time with facts. I mean, here we talk about loan and he bring up vault cash?! It is like he does not even know what bank is. He certainly does not know what balance sheet is. Par for the course for academic economics profession that embarrass itself daily in public view
Dc: whatever you do do not talk to people who run your local bank. It is better for your brain to live in scotts lala land where bank actually give a damn about reserve level before making loan. This is double true if you are academic economist and therefore unable to appreciate phlogiston
5. July 2010 at 20:39
Doc Merlin,
Regarding Turd Ferguson’s two points:
1. On this point, the cost of borrowing from the Fed to cover a reserve requirement might indeed factor into a bank’s decision making process in regard to making a loan, but this does not invalidate the notion that banks do not lend out of reserves per se. Even the BIS, the central bankers’ central bank recognise this. They noted in a recent paper that “the level of reserves hardly figures in banks’ lending decisions. The amount of credit outstanding is determined by banks’ willingness to supply loans, based on perceived risk-return trade-offs, and by the demand for those loans. The aggregate availability of bank reserves does not constrain the expansion directly.”
It is obvious why this is the case. Loans create deposits which can then be drawn upon by the borrower. No reserves are needed at that stage. Then, as the BIS paper says, “in order to avoid extreme volatility in the interest rate, central banks supply reserves as demanded by the system.”
The loan desk of commercial banks have no interaction with the reserve operations of the monetary system as part of their daily tasks. They just take applications from credit worthy customers who seek loans and assess them accordingly and then approve or reject the loans. In approving a loan they instantly create a deposit (a zero net financial asset transaction).
The only thing that constrains the bank loan desks from expanding credit is a lack of credit-worthy applicants, which can originate from the supply side if banks adopt pessimistic assessments or the demand side if credit-worthy customers are loathe to seek loans.
The extent of the bank’s lending decisions, then, is a function of how much credit risk the bank wants to take on at a given price (supply) and how much its customers wanted to borrow from us at a given price (demand) and never a function of the amount the bank had in its exchange settlement account (reserves for US readers) with the central bank.
2. The banks have a reserve requirement, but the Fed BY LAW has to provide the required reserves if the bank is short. It’s non-discretionary. If a bank fails to meet its reserve requirement it is subject to a reserve-deficiency charge. Reserve banks are authorized to assess charges at a rate of 2 percent above the discount rate. Thus, the Fed provides a third minor source of reserves for banks unable to acquire the required level of reserves from the fed funds market or the discount window, but it does provide the reserves.
So your “two fact” are actually incorrect.
6. July 2010 at 02:50
For the record, here’s an explanation of the facts from somebody who actually does understand the monetary system:
http://neweconomicperspectives.blogspot.com/2009/07/why-negative-nominal-interest-rates.html
http://neweconomicperspectives.blogspot.com/2009/07/why-negative-nominal-interest-rates_14.html
6. July 2010 at 04:14
“123, Where is the evidence that he has changed his mind?”
In this interview Hall is parroting Bernanke: “Depletion of the capital of financial institutions raises financial frictions to levels that distinctly impede economic activity”
6. July 2010 at 04:17
Marshall Auerback also understands the system very well, of course.
6. July 2010 at 06:44
Turd, You said;
“Way to change the subject. I didn’t ask why the bank wouldn’t buy a Treasury. I asked why the bank wouldn’t make the loan to you given the payment of interest on reserve balances.”
It makes no difference whether the loan is to me or the government. If you raise the interest rate on reserves, you raise the opportunity cost of lending that money out, whether to me or to the government.
You said;
“If the borrower is creditworthy, most banks should want to make the loan.”
Making loans is riskier than holding reserves, that’s why banks hold lots of ERs right now. The spread between the IOR and the prime rate must be large enough to compensate for the risk. The higher the IOR, the fewer loans that banks will want to make (ceteris paribus).
You said;
“And that was their point. It doesn’t matter if banks “hoard” reserve balances because they don’t need them to make loans and create deposits in the first place.”
If that was their point why didn’t they just say so? I have no interest in bank lending policy, merely monetary policy. Monetary policy is about the supply and demand for base money. A higher IOR increases the demand for bank reserves, and hence is deflationary. Bank lending mostly reflects the strength of the economy–the Fed controls the base (and the IOR) which drives NGDP, and bank lending responds endogenously to changes in NGDP.
Turd, Actually, the IOR has occasionally been above the target rate. And when it is higher than the fed funds rate, it is the IOR, not the fed funds rate, that is the opportunity cost of reserves.
Marshall, You said;
“On this point, the cost of borrowing from the Fed to cover a reserve requirement might indeed factor into a bank’s decision making process in regard to making a loan, but this does not invalidate the notion that banks do not lend out of reserves per se. Even the BIS, the central bankers’ central bank recognise this. They noted in a recent paper that “the level of reserves hardly figures in banks’ lending decisions. The amount of credit outstanding is determined by banks’ willingness to supply loans, based on perceived risk-return trade-offs, and by the demand for those loans. The aggregate availability of bank reserves does not constrain the expansion directly.””
There is no contradiction between the IMF position and my position. When did I say banks were constrained to make loans by their level of reserves? My point was very simple, if a bank wants to get rid of reserves it can—it can lend then out. Because banks do not control the size of the monetary base (I hope we can at least all agree on that point, which is M&B 101) then the only way banks can get rid of reserves, in aggregate, is by turning them into cash held by the public. That was my very simple point, and I would hope it is non-controversial.
123, I agree with that quotation as well, and I don’t see where Hall denied that in 2009.
Zanon, The worst way to learn about monetary theory is to talk to bankers. The core theorem of monetary theory (the public determines the real demand for money and the Fed determines the nominal supply of money) is founded on the fallacy of composition. Because bankers don’t see the big picture, they have no idea of the role they play in the system. Bankers can get rid of base money they don’t want. The public as a whole cannot, unless the Fed lets them
6. July 2010 at 06:47
Sumner: the worst way to learn monetary Theory is to talk to banker I agree. But they can tell you about monetary Reality
If you are uninterested in real world or fact rest assured I point people to clueless academics who do not understand reserve, balance sheet, or bank but does not stop them from opining on this
6. July 2010 at 07:05
Let me give example that your commentors can chew on. You should not read this Scott, it will burns!
Let’s look at real life loan — car loan, no ridiculous suitcase of cash example.
Person take out car loan to buy car. his bank credit receivable (loan) and debit reserve. Car salesman sell car. His bank credits liability (deposit) and credit reserve. Reserve transfer through normal daily interbank operations, or overnight market.
Note: 1. loan created deposit. 2. Reserve shifted from buyer bank to seller bank but were neither created nor destroyed at system level by transaction. 3. Reserve were not “loaned out”, they just use for settlement balances.
Also note — this makes mockery of scott sumner assertion that “banks lend out reserve” and can, at system level, get rid of reserve they don’t want by making loan.
6. July 2010 at 07:38
I find this discussion on whether banks need reserve balances to make loans pretty interesting.
Here’s something I would like to add:
@turd ferguson
“The bank buys the vault cash from the Fed and receives an overdraft to its reserve account, which it clears by the end of the day, probably through the fed funds market or other money markets.”
“Then the final tally is an increase in loans of 10K on the asset side and an increase in money market borrowings of 10K on the liability side.”
Turd, banks could indeed create loans that way, but I think whether the fund is obtained through fed funds market or other money market, it’s subject to reserve requirement, and hence total excess reserves in the banking system will decrease while required reserves will increase.
Hence there will come a point where excess reserves will all become required reserves, and you can no longer create any loans.
@Marshall Auerback
“The banks have a reserve requirement, but the Fed BY LAW has to provide the required reserves if the bank is short. It’s non-discretionary. If a bank fails to meet its reserve requirement it is subject to a reserve-deficiency charge. Reserve banks are authorized to assess charges
at a rate of 2 percent above the discount rate.”
Hmm, I wonder if Marshall could be more specific about the ‘LAW’, but here are some sentences from Regulation D on reserve requirement part 204.6 on the subject of charges for reserve deficiencies:
“Reserve Banks may waive the charges for reserve deficiencies except when the deficiency arises out of a depository institution’s gross negligence or conduct that is inconsistent with the principles and purposes of reserve requirements.”
“Violations of this part may be subject to assessment of civil money penalties by the Board under authority of Section 19(1) of the Federal Reserve Act (12 U.S.C. 505) as implemented in 12 CFR part 263. In addition, the Board and any other Federal financial institution supervisory authority may enforce this part with respect to depository institutions subject to their jurisdiction under authority
conferred by law to undertake cease and desist proceedings.”
If a bank simply create loans out of reserve deficiencies to make some ‘obscene profits’, then it’s probably a ‘gross negligence with the principles and purposes of reserve requirements’ and is punishable by Federal Reserve Act. Whatever that ‘LAW’ is, it seems to contradict with regulation D here.
And by the way, the charges is 1% above discount rate.
Here are some related links:
Regulation D
http://ecfr.gpoaccess.gov/cgi/t/text/text-idx?c=ecfr;sid=6fa3298dd1251a8435aa6e2ffb6fb5b4;rgn=div5;view=text;node=12%3A2.0.1.1.5;idno=12;cc=ecfr#12:2.0.1.1.5.0.2.6
Fed Funds
http://www.newyorkfed.org/aboutthefed/fedpoint/fed15.html
Reserve Requirement
http://www.newyorkfed.org/aboutthefed/fedpoint/fed45.html
6. July 2010 at 07:44
I forget where I read this comment, but it’s obvious enough not to require attribution: By paying interest on “reserves,” the Fed converted them from money into debt securities, the functional equivalent of T-bills. “Reserves” on which interest is being paid aren’t really part of the money supply. Instituting positive interest on “reserves” was, in effect, a massive OMO–a contractionary one. (Granted, one can get only so far attempting to deal with these matters using a crude absolute distinction between *money* and *non-money*.)
On a meta-topic: I wish Anon, Zanon, and Turd would adopt the positive tone of most of your commenters–focusing on the issues, omitting sneers *ad hominem*.
6. July 2010 at 07:51
Right, Zanon. Scott’s example doesn’t prove his point (as if a bank can make a profit on its asset-liability spread while earning 0.25% on reserve balances, or whatever reserve balances earn compared to liabilities–there are NO banks that can survive earning only the spread from money market rates on assets–and thus use this as a rationale to turn down credit worthy borrowers) and my counter example (essentially the same as Zanon’s example) is how lending actually occurs the vast majority of the time (the percentage of loans that end up in 100% currency withdrawals has to be way down in the single digits, I would imagine).
6. July 2010 at 07:56
Mark C,
Because the Fed is setting an interest rate target, it supplies reserve balances on demand to banks. If reserve requirements grow in the aggregate, the Fed accommodates through open market operations or else it can’t hit its interest rate target. There is no constraint on the banking system from reserve requirements. The only actual effect for an individual bank if its reserve requirement grows is that it might have to borrow in the money markets to cover this. This raises the costs for the bank, and thus perhaps the rate it would charge a credit worthy borrower, but it does not in any way constrain the bank’s ability to create the loan.
6. July 2010 at 08:07
@Turd:
From a macro perspective, because of the reserve requirement, the bank is limited in the amount it can loan by the reserve requirement. This isn’t a hard limit, as you said, but rather caused by increased financing costs. Never-the-less, in the aggregate it still behaves as a limit to money creation, even if no bank is constrained at the margin, by a hard limit.
6. July 2010 at 08:09
An interview with Robert Hall by the MN Fed:
http://www.minneapolisfed.org/publications_papers/pub_display.cfm?id=4452
6. July 2010 at 08:09
“If that was their point why didn’t they just say so? I have no interest in bank lending policy, merely monetary policy. Monetary policy is about the supply and demand for base money. A higher IOR increases the demand for bank reserves, and hence is deflationary.”
If you think banks lend out reserves, apparently you have an interest in lending policy, whether you know it or not.
The purpose of reserves is in their use as settlement balances in making interbank payment for net clearing and net settlement of customer transactions.
Banks do not “lend out reserves”. The only exception to this is the fed funds market, which is restricted to banks by design and purpose, and off limits to the rest of the transacting universe. Reserves are inaccessible to nearly all market participants, including nearly all borrowers and depositors who are involved in nearly all credit and money creation activity.
Reserves are a financial asset. Reserves earning IOR are equivalent to treasury bills. Both provide banks liquid access to settlement balances, reserves being one step closer to them.
Banks create and lend the money they extend in making a loan. They don’t “lend out reserves” in making a loan anymore than they lend out treasury bills in making a loan.
Banks do not make credit and lending judgments based on their reserve positions. They make them based on credit quality and capital availability. Therefore, nearly all money creation has virtually nothing to do directly with bank reserves.
Please read up on this at Post Keynesian websites and learn how actual banking practice factually supersedes your “monetary theory”.
6. July 2010 at 08:11
Scott says, “There is no contradiction between the IMF position and my position. When did I say banks were constrained to make loans by their level of reserves?”
If you agree they aren’t constrained, then what’s the problem with paying interest on reserve balances? Banks can hold their reserve balances and earn the remuneration rate AND make loans to any credit worthy borrower that comes through the doors.
Even if the loans result in a loss of some reserve balances through withdrawals, there’s no “opportunity cost” issue for reserve balances because making loans and creating deposits doesn’t use up reserve balances. Again, for the bank, the issue is one of portfolio management and the basic business model of banking, not opportunity cost; it can’t make a sustainable profit holding a bunch of reserve balances earning interest. It’s quite well known that banks need an average interest spread probably above 3% to be profitable, and remuneration on reserve balances alone will almost never provide that, and certainly isn’t providing anywhere near that now.
6. July 2010 at 08:14
MarkC,
Your responses to Marshall again miss the point I make in 6. If bank reserve requirements grow in the aggregate, then there is pressure in the federal funds market for the rate to rise above the target rate. The Fed then does an open market operation to add reserve balances and achieve its target.
6. July 2010 at 08:16
Doc,
It’s widely known that reserve requirements are a tax on banks, essentially. Nobody’s disputing that. But suggesting there’s an effect beyond that in a bank’s “ability” or “capacity” to create a loan is incorrect, and confuses reserve requirements with capital, as Anon noted.
6. July 2010 at 08:19
Philo,
On the tone, I agree and for my own part will try to be as congenial as Scott and other regulars here have been.
6. July 2010 at 08:32
Turd, u said:
“If bank reserve requirements grow in the aggregate, then there is pressure in the federal funds market for the rate to rise above the target rate. The Fed then does an open market operation to add reserve balances and achieve its target.”
If that’s the case, the Fed could simply raise the fed funds target rate and the IOR somewhere close to where the fed funds market is trading, then the pressure is being eased whithout having to add any reserve balances, and that’s exactly why Scott has been arguing that IOR is contractionary, because it takes away the incentive for banks to create loans.
6. July 2010 at 08:37
BTW, given the actual functional role of reserves, your proposal for a negative IOR rate has never held together as a matter of logical construction.
IOR will always be a function of the fed funds target rate.
So what you’ve been proposing in effect is a FFR target that is sufficiently negative to accommodate the negative IOR you desire.
The negative IOR rate itself is a side show in that context. The full effect on economic activity and banking system behavior is determined by the negative FFR target.
6. July 2010 at 09:13
“If that’s the case, the Fed could simply raise the fed funds target rate and the IOR somewhere close to where the fed funds market is trading, then the pressure is being eased whithout having to add any reserve balances,”
Well, the Fed will raise the rate if it perceives economic activity improving, for sure. But the effect is with a lag, as the reserve requirement increase has already happened.
Also, note that with sweep account technology, reserve requirements are essentially optional anyway. Look at reserve requirements over the past 12 years or so and you will see almost no increase (until fall 2008, when the benefit of sweeping balances declined) even as lending obviously continued to grow (aside from recessions).
“and that’s exactly why Scott has been arguing that IOR is contractionary, because it takes away the incentive for banks to create loans.”
This doesn’t follow from your first point. The only reason to raise the rate in the first point from the Fed’s perspective is to slow the economy down because it’s growing too fast. In that case the Fed wants to reduce the incentive for loan creation. Scott’s argument about IOR is that it reduces the incentive for loan creation in the opposite case, when the Fed wants more loan creation.
6. July 2010 at 12:15
Doc & Philo: I will show Sumner the same respect you extend to flat earthers when you try to convince them that planet is, actually, round.
It is not like this stuff is hard. And look at extreme arrogance of this comment: “The core theorem of monetary theory (the public determines the real demand for money and the Fed determines the nominal supply of money) is founded on the fallacy of composition. Because bankers don’t see the big picture, they have no idea of the role they play in the system.”
I hates bankers. But Sumner and his ilk hold themselves up as people too smart of “fallacy of composition” who can “see big picture” and then they display knowledge of basic loan as taking out suitcase of vault money from bank and eating it with ketchup (or something, certainly it is not used for actually buying something on planet earth).
If Sumner was to apply his “fallacy of composition” “big picture” brain to, say, human reproduction, he would say babies come from hospital and start writing papers on how stork migration patterns impact population growth. It is imbecilic.
Banks, at system level, do not lend out reserves. They simply do not. Therefore, quantity of reserves do not impact whether banks make loans or not. OIR has no impact on bank lending.
I have shown very clearly how, in case of simple car loan, reserve is debited at one bank and credited at another bank by exact same amount. This blows great smoking hole in “banks lend reserve” at system level. It leaves entire Sumner position as smoldering wreck. It is not first time this has been pointed out to Sumner, and he is lost cause for obvious reason, but you as commenter maybe interested in learning something new should be able to smell the stinky fish.
Reserve requirement, and fractional reserve banking, impose no limit hard or soft on bank lending at system level. Canada has requirement of zero. Capital requirement pose hard limit. Academic economist do not understand this fundamental difference — no doubt because they are too focused on “big picture” or navel. Your guess.
MarcC: Fed has option of doing its job and maintaing FFR at target, or blowing up interbank market. If system hits reserve constraint, which it can for any reason for technical only, those are its choices. Do its job, or blow up.
I mean hecks. Enough people develop appetite for Scott Sumner vault cash with ketchup, and walk out with suitcase of money from deposit account, system could hit reserve constraint (and not there has been no lending). Ben Bernanke says “oh no, do I do my job or let system blow up?! People want to eat dollar bill with ketchup! Let me call Scott Sumner so he can tell me whether I should act normal or destroy interbank market, and with it, banking system”.
Scott sumner will write post on nominal tomato.
Brilliant.
6. July 2010 at 12:32
excellent comment, zanon
7. July 2010 at 05:12
I think that monetarist/creditist divide goes like this – monetarists think that private sector assets should be bought by the fed only in very exceptional circumstances when zero bound has been reached, while creditists support such interventions by looking at credit spreads irrespective of the level of short term interest rates. By saying that BAA corporate bonds are undervalued and purchases of BAA corporate bonds should be considered by authorities, Hall has moved a bit toward creditist end of the spectrum, although it is not very clear if he went as far as Woodford did in his Bank of England presentation.
P.S. I am going on a holiday (without internet access), so I won’t be able to see your reply during next 14 days.
7. July 2010 at 07:18
zanon, you said: “Person take out car loan to buy car. his bank credit receivable (loan) and debit reserve. Car salesman sell car. His bank credits liability (deposit) and credit reserve. Reserve transfer through normal daily interbank operations, or overnight market.”
“I have shown very clearly how, in case of simple car loan, reserve is debited at one bank and credited at another bank by exact same amount. This blows great smoking hole in “banks lend reserve” at system level.”
Unfortunately, the “great smoking hole” in this story is that there are now more deposits in the system, which requires additional reserves at the system level.
Now, as a general rule the Fed endogenizes reserves to maintain its FFR target and this seems to be the source of the ongoing meme that reserves don’t matter. But, if, other things being equal, the Fed decides to change its reserve policy it must change the FFR level required to retain the same amount of credit outstanding. Paying interest on reserves raises the opportunity cost of lending, which will result in a contraction of lending without a compensating reduction in the FFR (or some other policy).
Anyway, there is plenty written on this subject. A quick google returned this: http://www.jstor.org/stable/4538081?seq=1
7. July 2010 at 07:19
Scott, I thought Hall’s thoughts on QE were interesting:
“Hall: First of all, I believe you should think of the Fed as simply part of the federal government when it comes to the financial side of its interventions. If you look at how the federal government responded initially, it was the Treasury that was providing the funds. Of course, TARP [Troubled Asset Relief Program] was there using the taxpayers’ money without involvement of the Fed. Also, early in the crisis Treasury deposited hundreds of billions of dollars at the Fed, which the Fed then used to buy assets. So there the Fed was just an agent of the Treasury. It was as if the Treasury took its funds to a broker.
Eventually, the Treasury was impeded from doing that by the federal debt limit. But the debt limit doesn’t apply to funds borrowed by the Fed, so it then started borrowing large amounts from banks by issuing reserves. That is what caused all the confusion about thinking this was somehow part of conventional monetary policy.
I would distinguish between conventional monetary policy which sets the interest rate and this kind of financial intervention of buying what appear to be undervalued private securities. Issuing what appear to be overvalued public securities and trading them for undervalued private securities, at least under some conditions and some models, is the right thing to do. In my mind, it doesn’t make a big difference whether it’s done by the Federal Reserve, the Treasury or some other federal agency.”
7. July 2010 at 07:47
Zanon, I have no problem with your car loan example. But no, it doesn’t refute my argument that banks can get rid of reserves they don’t want.
MarkC. I agree that the Fed tries to make it easy for banks. But nevertheless, they maintain control over the total amount of base money in circulation. It doesn’t seem that way as the base responds endogenously as needed to hit the central banks short run (interest rate) and long run (inflation) goals.
Philo, That’s a good point. I have no trouble with people thinking about interest-bearing reserves as bonds, or as cash where the normal QT models no longer apply because monetary policy is conducted through changes in the demand for money.
You said;
“On a meta-topic: I wish Anon, Zanon, and Turd would adopt the positive tone of most of your commenters-focusing on the issues, omitting sneers *ad hominem*.”
I have noticed a positive correlation between the politeness of commenters and the quality of their observations.
Turd, You keep talking about profitable business loans, but aren’t banks currently holding lots of low yield Treasury debt? It’s no wonder the interest-bearing reserves seem attractive. When T-bill yields were much higher, ER levels were very low.
Thanks Dilip.
anon, You said;
“Banks do not “lend out reserves”. The only exception to this is the fed funds market, which is restricted to banks by design and purpose, and off limits to the rest of the transacting universe.”
You are putting words in my mouth. I never claimed most loans are made for cash, nor do I believe that. But surely there are a few cases of someone borrowing a few thousand bucks and then immediately withdrawing that much cash from the bank, and then the bank replenishing cash out of deposits at the Fed. My only point was that banks could get rid of ERs if they wanted to. That’s all.
You said;
“Banks do not make credit and lending judgments based on their reserve positions.”
If you think I disagree with you then you haven’t understood a thing I have said. Reserves and loans are determined by separate factors. If a bank wanted to get rid of reserves, and didn’t find a good borrower, they’d usually just buy T-bills.
I’ve already wasted enough of my life reading Post-Keynesian economics, which I regard as the least appealing form of Keynesianism. So no, I don’t plan to read any more.
Turd, You said;
“If you agree they aren’t constrained, then what’s the problem with paying interest on reserve balances? Banks can hold their reserve balances and earn the remuneration rate AND make loans to any credit worthy borrower that comes through the doors.”
You are confusing money and credit. The recession was (primarily) caused by tight money, not tight credit. IOR makes money tighter, I never said the main macro problem we face is that IOR makes credit tighter. Perhaps it currently raises rates for borrowers by a quarter point, but the Fed funds target range is 0% to .25%, so even that isn’t clear. But ceteris paribus, a higher IOR means tighter money–that was my only point.
anon, The only purpose of a negative rate on ERs is to get rid of ERs. I am under no illusion that it would material affect other rates in the economy. The current problem is not that interest rates aren’t low enough, it is that the demand for money is too high, and hence the price of goods is too low.
123, Thanks. There may have been a shift in emphasis, but I doubt whether Hall has abandoned the monetary model he used throughout his entire life.
Enjoy your holiday.
7. July 2010 at 08:02
“Paying interest on reserves raises the opportunity cost of lending,”
Wrong. Wrong. Wrong.
Let’s take an example. Assume a bank with the following balance sheet:
Assets–100 reserve balances (rbs), 500 loans (l)
Liabilities/Equity (L/E)–500 deposits (d), 100 equity (eq)
Let’s assume reserve requirements are 10% and that capital requirements are 10%, for simplicity.
Assume that a creditworthy borrower gets a loan of 100, which creates a deposit of 100. Now the balance sheet looks like this:
Assets–100 rbs, 600 l
L/E–600 d, 100 eq
The bank is meeting both capital and reserve requirements.
Now, what if the recent borrower withdraws 100? The balance sheet looks like this:
Assets–0 rbs, 600 l
L/E–500 d, 100 eq
The bank is deficient in reserve requirements now, though not in capital. Let’s assume it would like to return to its previous level of 100 rbs and thus borrows in money markets. The balance sheet now looks like this:
Assets–100 rbs, 600 l
L/E–500 d, 100 borrowings (b), 100 eq
Both reserve requirements and capital requirements are again met.
For this bank, what is the marginal effect of making a loan? The answer is NOT the IOR, but the spread the bank can earn on the rate it earns on the loan and the rate it has to pay on its new liabilities. And if the bank can find more core deposits to replace the new borrowings, all the better. Indeed, if anything, the IOR are adding to bank profits compared to no IOR, and thus raising capital available to expand assets.
What if we assume instead that the bank starts out with no excess reserves (ER)? Assume the balance sheet started out like this:
Assets–50 rbs, 550 l
L/E–500 d, 100 eq
The bank makes a loan for 100, which creates a deposit of 100. But the bank, to meet its reserve requirement, also borrows 10 rbs in the money markets. The balance sheet loos like this:
Assets–60 rbs, 650 L
L/E–600 d, 10 borrowings, 100 eq
Both reserve requirements and capital requirements are now met.
Note, however, that the marginal effect of the reserve requirement is to raise the cost of the loan to the bank. But ALSO, note that the effect of any IOR is to REDUCE this tax from reserve requirements. Again, the IOR effect is NOT to reduce the incentive to lend. If anything, it’s the opposite.
Also, in the real world, it’s unclear that the loan would raise reserve requirements anyway, since (as just one example) the lender can use sweep account technologies. So, Truth’s point about more deposits and thus more requirements doesn’t stand up to actual experience of the past 15 years. (And, again, the comment about the Fed changing the FFR is off-target, as that implies the Fed sees increased economic activity sufficient to warrant a rate hike.)
Finally, if the borrower withdraws the 100 d as in the first example, the result that the IOR didn’t reduce the incentive to lend is unchanged (and, again, if anything helped compared to no IOR).
7. July 2010 at 08:04
“Turd, You keep talking about profitable business loans, but aren’t banks currently holding lots of low yield Treasury debt? It’s no wonder the interest-bearing reserves seem attractive. When T-bill yields were much higher, ER levels were very low.”
It’s common for banks to increase holdings of treasuries during recessions. By definition, there are fewer credit worthy borrowers. That in no way contradicts my point.
7. July 2010 at 08:11
And, of course, it’s those “low yielding treasuries” that are about the only asset class that earned a positive return in the first half of the year. Allen Sloan just wrote the following yesterday:
“The only investment that showed a double-digit positive return for the first half of this year was … long-term Treasury securities. Their total return — price gains plus interest — was 13.2%.”
7. July 2010 at 09:04
turd feguson said “Note, however, that the marginal effect of the reserve requirement is to raise the cost of the loan to the bank. But ALSO, note that the effect of any IOR is to REDUCE this tax from reserve requirements. Again, the IOR effect is NOT to reduce the incentive to lend. If anything, it’s the opposite.”
Paying interest on reserves makes the reserve requirement non-binding by making interest on reserves an attractive investment in their own right and, hence, other assets (including loans) relatively less attractive.
“Also, in the real world, it’s unclear that the loan would raise reserve requirements anyway, since (as just one example) the lender can use sweep account technologies. So, Truth’s point about more deposits and thus more requirements doesn’t stand up to actual experience of the past 15 years.”
I’m working in the confines of your toy example. “Sweep account technologies” reduce the cost of maintaining reserves but do not eliminate them. The point is that everything has a price and if you exogenously raise or lower the price of one asset (in this case, reserves), there will be spillovers into other assets. Now you might argue that the effects of paying IOR will be confined to shorter term low risk assets like Treasuries and won’t affect lending, but that’s a different argument.
7. July 2010 at 09:08
My first para was unintelligible it should have read:
Paying interest on reserves has made the reserve requirement non-binding and made reserves an attractive investment in their own right and, hence, other assets (including loans) relatively less attractive.
7. July 2010 at 09:28
“Paying interest on reserves makes the reserve requirement non-binding by making interest on reserves an attractive investment in their own right and, hence, other assets (including loans) relatively less attractive.”
That’s garbage. I already showed you how the loan itself is judged by the corresponding liability. And the IOR adds to profits and thus capital. Any response to that?
7. July 2010 at 09:31
“Paying interest on reserves has made the reserve requirement non-binding and made reserves an attractive investment in their own right and, hence, other assets (including loans) relatively less attractive.”
Again, I showed you that the loan itself is not competing with the rbs earning interest. It is competing with the return on the corresponding liabilities.
7. July 2010 at 10:37
“That’s garbage. I already showed you how the loan itself is judged by the corresponding liability. And the IOR adds to profits and thus capital. Any response to that?”
What is the corresponding liability to the loan? At the margin, it is some weighted average cost of debt (deposits, short and long term funding sources) and equity. Paying interest on reserves will raise the banks cost of short-term debt (and perhaps to much lesser extent longer term debt and equity) because some of the demand for bank debt will fall as a result of increased demand for reserves (holding reserves and lending in short-term markets are close substitutes). That higher cost of funding reduces the incentive to make loans.
7. July 2010 at 14:27
thruth: Yes, loan create deposit. So bank make loan, it create deposit. Reserve move from first bank to second bank. First bank does not care what reserve level is when it makes loan because bank lending is not reserve constrained, it is reserve indifferent.
Reserve move between bank all the time. Reserve is for payment settlement primarily.
So suppose system has a whole is short reserve. This can also happen if people take out cash and eat it with ketchup which is Sumner view of what a loan is. So reserve position is independent of loan, loan creation is just one factor. The system can ONLY now get reserve from discount window. Please note this is “system thinking” in action, something that academic like Sumner think they do when they say babies are grown in hospital.
So Fed has choice — either pay at discount window or have overnight interbank lending market break down, and throw FFR into tailspin.
Bank have many ways to balance reserve across system, ON IB is just one. But, if system is short (which may have nothing to do with loan) Fed can either do its job or destroy banking system. I know Sumner find this point very confusing.
OIR has no impact on lending. To the extent it recapitalizes banks, it is stimulative as bank lending is capital constrained.
TURD: You are absolutely correct in all of this. I am afraid that balance sheet will go right over their head.
Scott Sumner: You do not understand basic bank operation. You do not know what balance sheet is. You do not know what reserve is. Unfortunately for you, basic knowledge of this it out there and it makes you, and all other academic economics, looks like fools. I am sorry but it is truth. Earth is round and revolve around sun, no matter what Greg Mankiw might write in this book.
I have shown how bank making loan merely transfer reserve from A to B, it is not reduce reserve at SYSTEM LEVEL (which is where you claim your thinking is based, despite obvious evidence to contrary).
Banks can of course get rid of reserve they don’t want by buying t-bills. Your point is that bank ability to lend is somehow dependent upon their reserve and it is not. Simple bank operation prove this. Simple bank regulation prove this. Simple balance sheet prove this. Simple 5 minute discussion with anyone at bank prove this. Simple example prove this. Open up a paper of your choice, look at lending, look at unemployment, look at reserve level, look at inflation rate, and the answer stare everyone in face.
It is your cognitive block that make you unable to see this plain truth. There is no hope for you.
Anyone else who has not swallowed cool-aid cannot help but to see how loans create deposit, not other way around, and bank lending is not capital constrained, and reject your party line as nonsense.
Bank at SYSTEM LEVEL cannot get rid of reserve by making loan. It just go to another bank, unless they dine on Scott Sumner dish of ketchup and vault cash.
Entire “money multiplier” model, and monetary theory in general, is based on absolutely fallacious mechanism.
7. July 2010 at 14:30
Just so I am perfectly clear. If system is short reserve, checks start bouncing even if accounts have enough money. Payment system itself breaks down if Fed decides to close up shop and go fishing.
7. July 2010 at 15:53
“You are putting words in my mouth. I never claimed most loans are made for cash, nor do I believe that.”
You don’t understand the point I’m making at all. I’m not even talking about cash reserves.
“My only point was that banks could get rid of ERs if they wanted to.”
No they can’t. This is hyperbolically divorced from reality. Banks can’t force cash on their customers and they can’t “get rid” of $ 1 trillion currently in the system. A mere passing familiarity with the average effective reserve ratio and pro forma balance sheet implications would demonstrate that.
“I’ve already wasted enough of my life reading Post-Keynesian economics, which I regard as the least appealing form of Keynesianism”
That’s understandable. Its factual basis probably makes it the one that’s most orthogonal to your theories.
7. July 2010 at 15:55
above meant $ 1 trillion in reserve deposit balances with the Fed
7. July 2010 at 16:39
anon is, of course, exactly right.
But in Sumner’s reality, banks serve up vault cash with ketchup with households then eat while the Fed shuts up shop to go fishing.
I laugh it is so funny!
7. July 2010 at 16:44
‘Banks can’t force cash on their customers and they can’t “get rid” of $ 1 trillion currently in the system.’
They could purchase assets with that money. In a less regulated banking system, they would probably purchase equities or stakes in companies, much the way that insurance companies do now. The rules for banks now tend to heavily favor debt over other type of assets however, which in my opinion makes the system brittle.
‘Your point is that bank ability to lend is somehow dependent upon their reserve and it is not. ‘
We never said that it was dependent on a reserve for an individual bank. The capitalization requirements do limit an individual bank, but the reserve requirements just serve to raise the costs to the bank and thus the rates that the bank lends out. This limits lending because of consumer demand for debt at that rate.
What I have said is that /in aggregate/, total lending for banks is limited by base money INPART because of the costs imposed by reserves. In countries without reserve requirements, liquidity risks do play this role (as you said earlier).
I think the core of the disagreement between you and Scott is that he thinks that a small tax on reserves would increase lending (or atleast result in banks switching from federal reserve reserves to buying assets of some sort. You don’t seem to think it would?
7. July 2010 at 17:12
zanon said “Just so I am perfectly clear. If system is short reserve, checks start bouncing even if accounts have enough money. Payment system itself breaks down if Fed decides to close up shop and go fishing.”
The other option is that the banks entice customers out of reserve required deposits into non-reserve required accounts. They would probably have to sell some assets to do it and there’ll be a lot of pain and anguish. Private arrangements would eventually spring up in the absence of a central bank.
btw, please tone down the ad homs.
7. July 2010 at 17:21
“They could purchase assets with that money.”
They can purchase assets all they want, but it has no bearing on the amount of reserves in the system.
Bank customers can effectively drain bank reserves by demanding currency. But there’s not nearly enough natural demand for currency to drain $ 1 trillion in excess reserves.
Banks can gradually convert excess reserves to required reserves by acquiring new assets and creating reservable deposits in the process. But their balance sheets and capital now or prospectively won’t be remotely big enough to convert $ 1 trillion in this manner.
7. July 2010 at 17:51
@anon
‘Banks can gradually convert excess reserves to required reserves by acquiring new assets and creating reservable deposits in the process. But their balance sheets and capital now or prospectively won’t be remotely big enough to convert $ 1 trillion in this manner.’
Well, they could it will just take a while. USD1T is a hell of a lot of money.
7. July 2010 at 17:58
“they could it will just take a while”
That’s absurd.
Current required reserves are about $ 50 billion.
The banking system would have to expand 20 times in size in order to absorb it all.
Nothing personal, but this is a typical reality-divorced observation emanating from “monetary theory” as practiced on this blog.
7. July 2010 at 18:42
thruth: And what non-reserve assets precisely would banks entice customers into? Treasury bill? How would banks do this? T bill rate is already so low it is hard for them to limbo under. Maybe if Fed raised FFR to 10% then people would move to t-bill and out of reserve. PROBLEM SOLVED except now you have to admit that low interest rate is CONTRACTIONARY. DOH!
Suppose banks offer special on toasters, so depositers buy the toaster. Well, the toaster maker must be paid for toaster, and what does he do with money? Eat it with ketchup? No, he put it in bank and BANG it is reserve again.
Suppose bank say they give you $2000 if you buy stock. OK, you buy stock, but someone sell stock, and put cash in bank and BANG reserve exist again! Those darn reserve!
Maybe bank have special on ketchup and run ad campaing on how delicious ketchup is with vault cash. The Sumner Solution!
Have you lost sight on why reserves brought up in first place? It is because of ridiculous, ludircious, nonsense macroeconomic fiction of “money multiplier” where reserve is somehow constraint in bank lending, and extra reserve is somehow hot potatoe burning hole in bank pocket.
anon: “monetary theory” as practiced on this blog is absolute same “monetary theory” as practiced in actual pretigeous university like Harvard. It smell same no matter where you sniff.
7. July 2010 at 18:57
@all parties left standing:
On the $1tr in outstanding reserves, let’s not forget that the Fed used the proceeds to buy up a similar amount of MBS. The Fed would presumably have to sell back a similar amount of assets and drain the reserves to undo the IOR policy.
In the interview, I think that Hall basically concedes that the IOR policy wasn’t really monetary policy because it was simply an exchange of assets (reserves for MBS and agency debt):
“Eventually, the Treasury was impeded from doing that by the federal debt limit. But the debt limit doesn’t apply to funds borrowed by the Fed, so it then started borrowing large amounts from banks by issuing reserves. That is what caused all the confusion about thinking this was somehow part of conventional monetary policy.”
So looking at the policy as a whole it’s hard to say whether there was any monetary contraction.
All that said, if the Fed were to now decide to make the rate on ERs negative without draining the reserves, ERs would become a hot potato. Since there is no way to get rid of them at the aggregate level, why wouldn’t the incentives associated with negative ERs drive banks to raise capital and increase lending? The promise of new lending should help to raise NGDP expectations.
7. July 2010 at 19:10
zanon: “thruth: And what non-reserve assets precisely would banks entice customers into? Treasury bill? How would banks do this? T bill rate is already so low it is hard for them to limbo under. Maybe if Fed raised FFR to 10% then people would move to t-bill and out of reserve. PROBLEM SOLVED except now you have to admit that low interest rate is CONTRACTIONARY. DOH!”
You seem to be confusing assets and liabilities. The deposit, a liability, carries a reserve asset requirement. In our hypothetical, the system does not have enough reserve assets for all of the deposits. If the Fed were to refuse to provide more reserves, the solution is that the banks must entice depository customers into a non-reserve required account (e.g. a money market fund) by offering a sufficiently attractive interest rate to compensate for the extra risk the customer must bear.
7. July 2010 at 20:03
thruth: Please. I am meticulous about asset/liability difference.
all monetary policy is asset swap. My gods, look at how Fed sets FFR! It swaps one Fed liability/private asset for another Fed liability/private asset. I have found that in every discussion with “monetarists” they redefine the term to claim that fiscal policy (which actually change non-Govt net financial asset) is actually monetary policy. The last refuge of scoundrels.
In your hypothetical where system is short reserves overall you have checks bouncing right left and center because payment settlement mechanism has broken down. Bank is not interested in making loan because is too focused on not being burnt by angry mob. They may also be trying to track down Fed which, instead of doing job, has gone fishing. The situation you describe is like talking about monetary policy in the instance where earth is being demolished by meteor from outerspace and president has decided to take nap instead of flipping “stop meteor” button.
And in this world, where banks are gripped with illiquidity and are going out of business left and right, your brilliant marketing plan is “customer, please move money out of FDIC insured account into an uninsured account”. I can only guess and quantities of toaster and ketchup bank would have to offer idiot customer from him to do such a riduclous thing.
But that you for clarifying your belief that the money market fund, a foul and base pestilence that should be eradicated once and for all, is in your mind the SAVIOR in the event of liquidty crises and interbank market breakdown. That is of course the exact opposite of what happened not 12 months ago, but if you’ve ignored reality to date, why stop?
8. July 2010 at 04:16
zanon: that’s all beside the point. You gave a contrived example, I gave a contrived response.
You didn’t address this:
“If the Fed were to now decide to make the rate on ERs negative without draining the reserves, ERs would become a hot potato. Since there is no way to get rid of them at the aggregate level, why wouldn’t the incentives associated with negative ERs drive banks to raise capital and increase lending?”
8. July 2010 at 05:47
thruth: are you kidding me? Sumner characterized loan as man walking out of bank with suitcase of vault cash to buy nothing! You characterized system where fed was gone fishing while it was short reserve and payment system breaking down!
I show you how in course of normal payment settlement (and credit creation) reserve is maintained within system. Bank at system level cannot get rid of this buy lending.
If Fed has tax on ER, then bank swap reserve for Treasury bill. Maybe you say that Govt is no longer issuing Treasury bill either. They certainly would not lend to deadbeat while economy still weak.
Bank would need to raise capital as they are now being taxed on reserve and need to keep some reserve for regulatory reason, and tax is draining capital. Quality of loan demand is unchanged (miserable). This would not spur lending for reasons obvious to ayone who understand reserve, and by extension, actual monetary system.
8. July 2010 at 06:03
zanon, you said “I show you how in course of normal payment settlement (and credit creation) reserve is maintained within system.”
Noone is disagreeing that the Fed generally provide reserves to maintain the payment system. The disagreement is about what happens when the Fed changes its policy. Obviously shutting down the Fed if the system was reserve constrained would result in a panic. Otherwise it would simply put a constraint on deposit growth, marginally raising banks’ cost of finance.
“If Fed has tax on ER, then bank swap reserve for Treasury bill. Maybe you say that Govt is no longer issuing Treasury bill either. They certainly would not lend to deadbeat while economy still weak.”
Supply and demand set the price for Treasuries. If supply remain constant, the tax on ERs will raise the price of Treasuries. This provides the incentive for banks to raise capital and lend.
“Quality of loan demand is unchanged (miserable). This would not spur lending for reasons obvious to ayone who understand reserve, and by extension, actual monetary system.”
You are arguing there is no price at which banks would lend. If the return on other alternatives is reduced, why doesn’t the marginal incentive to lend increase? You can argue the incentive response is weak, but zero?
8. July 2010 at 07:19
Turd, Your example of 13% return on T-bonds is of course irrelevant, as it is the ex ante returns that matter. Ex ante returns on T-securities are very low, even for longer term Treasuries.
Regarding opportunity cost of reserves: Suppose the Fed raises the IOR to 23%. Would banks rather lend out money to business borrowers, or park the money in safe reserves at 23%. That’s why I think the IOR is the opportunity cost of using reserves for some other purpose. I just can’t see a bank lending money to a business at 8%, when they could lend to the Fed at 23%.
Zenon, You said;
“Scott Sumner: You do not understand basic bank operation. You do not know what balance sheet is. You do not know what reserve is. Unfortunately for you, basic knowledge of this it out there and it makes you, and all other academic economics, looks like fools. I am sorry but it is truth. Earth is round and revolve around sun, no matter what Greg Mankiw might write in this book.”
1. They why debate me?
2. Even if you are right, how would that impact my understanding of monetary policy. I also know little about drug lords, and yet during normal times they hold as much base money as banks.
You said;
“Bank at SYSTEM LEVEL cannot get rid of reserve by making loan. It just go to another bank, unless they dine on Scott Sumner dish of ketchup and vault cash.”
If the Fed put a negative 3% charge on ERs (including vault cash), believe me, they’d get rid of reserves really quick.
anon, See my answer to zenon. If banks don’t want ERs, then buy something with it. If banks are afraid the money will be redeposited in the banking system, they could put a negative 10% rate on bank deposits. This isn’t rocket science.
8. July 2010 at 20:41
scott: I am not debating you because you are in unfortunate position of learning wrong trick when young, and now you are old dog. I do bring up your concept of “loan” because it is illuminating and hysterical. it is joy when education and comedy come together so nicely.
thruth: what Fed policy are we talking about? Shutting down payment settlement infrastructure and destroying banking system by not addressing system level reserve shortage is only considered a “policy” option by academic macroeconomists. So, if Fed wants to not destroy country it *must* make up any reserve shortfall. I have no interest in further entertaining this obviously moronic fantasty but feel free to write about it in academic journal. include pictures of bullets, cannibalism, etc.
Banks lend when marginal revenue exceed marginal cost, same as any other business. Marginal revenue is marginal credit worthy customer. marginal cost is marginal cost of capital which must be set aside for that loan. Please note that reserves have nothing to do with this. Please also note how marginal credit worthy customer is very different in boom vs bust. Banks are pro-cyclical, which is why trying to get them to act counter-cyclically reveals profound ignorance.
You can order bank to make unprofitable loan (loan which will not be paid back) but now that is fiscal policy and this is commonly done in China, and in some parts of US too.
In general, aside from totemic pagan worship of academic macroeconomics, reserve is vertigal and unimportant part of financial system. So some fools want to genuflect in front of appendix, who cares? However, now you have me thinking. Any deposit in banking system credit reserve. So, if you want system wide reserve debit money must leave banking system entirely, either walking out the door as suitcase of “vault cash” (the “sumner loan”) or going back to Government in form of Treasury.
Even your beloved non-FDIC insured money market has CP loan somewhere, and what did corporation do with that loan? They did not take it out as vault cash, add ketchup, and eat is as Scott Sumner things. It almost certainly lives as some deposit somewhere, you gots it, crediting reserve.
ANON or TURD or anyone who actually has a clue and might be on this board, am I right? Can money stay in banking system and not create positive reserve credit? Is only way to get system debit for money to leave banking system altogether via cash of Treasury?
8. July 2010 at 22:59
zanon,
you are definitely right
SS,
“This isn’t rocket science”
No it’s not. The problem is that your academic pride of being ignorant of banking mechanics such as lending means among other things you do not understand the monetary system.
e.g.
“If banks are afraid the money will be redeposited in the banking system”
That alone is an absurd statement as a qualification – that money MUST be redeposited in the banking system, by some non-bank customer of a bank. That’s the only way the payee can get value for the bank check/draft or whatever the liability form is.
Similarly, paying negative interest on deposits can’t get rid of bank reserves.
The only way bank reserves can disappear is by public currency demand or active Fed intervention.
I’ve already explained that the former is a silly scenario, because it would require a doubling of publicly held currency. Your idea I suppose is to charge interest on deposits while leaving currency zero interest bearing, thereby increasing the demand for it. That’s a ridiculous tangent. Apart from being mechanically crazy, why would a central bank expect the public to spend currency to some monetary effect when it has just increased the real interest rate on it?
And the only way excess reserves can be converted to required reserves is for a massive 20X expansion of the US banking system balance sheet. That again is an absurd scenario.
Finally, look to the example of a zero reserve system like Canada for the final evidence that your theory is full of obvious holes, prima facie, without laboring through the mundane obviousness of the detail in the US case.
8. July 2010 at 23:02
above, meant relative real interest rate on currency
8. July 2010 at 23:03
i.e. compared to deposits, duh
9. July 2010 at 02:16
“Finally, look to the example of a zero reserve system like Canada for the final evidence that your theory is full of obvious holes, prima facie, without laboring through the mundane obviousness of the detail in the US case”
Canada isn’t a zero reserve system. It is a zero REQUIRED reserve system.
@Anon:
from your own link
“So they will try to trade them in the interbank markets. But in the aggregate, banks only trade the existing quantity of balances among themselves as only a change in the central bank’s balance sheet alters the quantity of balances circulating. So unless the central bank takes action to drain the reserve balances, the undesired excess quantity will just lead banks to bid the overnight rate down to the rate paid on reserve balances”
Sigh… thats the entire point. Weather you talk about it from a multiplier methodology or from a rate methodology it doesn’t matter. Interests paid on reserves are a floor on the overnight rate. The rate needs to drop to a more negative real rate, one way to do this is to do what sumner described. I think you guys are way too hung up on the terminology because of the post keynesian obsession with rates and completely missing the point. Th e fed charging interest on reserves will force the real rate down, allowing people to re-fi and acting like a monetary stimulus.
For the record I think the fed shouldn’t charge interest or pay interest on reserves.
9. July 2010 at 02:19
@anon:
This is untrue.
‘That alone is an absurd statement as a qualification – that money MUST be redeposited in the banking system, by some non-bank customer of a bank.’
No, it doesn’t have to be. It can be withdrawn from the banking system and stored as foreign currency reserves by a foreign government.
9. July 2010 at 05:46
zanon, you said “thruth: what Fed policy are we talking about? Shutting down payment settlement infrastructure and destroying banking system by not addressing system level reserve shortage is only considered a “policy” option by academic macroeconomists.”
You were the one providing the over-the-top example about shutting down payment settlement, not me. I just tried to demonstrate that, contrary to your assertion, it wasn’t necessarily true the unwillingness of the Fed to provide reserves on a given day does not send the whole economy over the brink. Of course, without an adequate explanation, I’m sure the economy would respond negatively to the news.
In any case, you don’t have to listen to me, listen to a Senior Economist in the Open Market Operations Department at the FRB NY (this is the Jstor article in the Journal of Post Keynesian Economics that I linked to earlier, that you obviously didn’t bother to look at…) commenting on a Post Keynesian paper making your claim:
BEGIN QUOTE
The author argues […] that “The Federal Reserve has no choice but to accomodate and provide all … required and excess reserves demanded.” In fact, the Federal Reserve could refuse to provide the reserves. While it would be an unlikely policy, the Federal Reserve could close the discount window and not meet the full demand for reserves through open market operations. If it did so without prior warning, banks would fail to meet their requirements and would be charged a penalty. That would probably change their behavior quickly. If the banks were given advance warning that the Federal Reserve would no longer accomodate reserve demands in excess of targeted levels through either open market operations or the discount window, banks would work to shrink required reserves so as to build up excess reserves to meet future reserve needs.
Under operating procedures that have been pursued by the Federal Reserve in recent years, a decision not to provide nonborrowed reserves in the amounts demanded by the banks would work through to slower money growth in a few weeks. Banks are restricted as to frequency and amount of discount window credit, so policies that force higher borrower will tend to raise short-term interest rates as banks use up their welcome at the window.
END QUOTE
So this would clearly lead to a contraction of supply, raise banks cost of lending and slow economic activity.
But let me repeat, this is a ridiculous example.
“Banks lend when marginal revenue exceed marginal cost, same as any other business. Marginal revenue is marginal credit worthy customer. marginal cost is marginal cost of capital which must be set aside for that loan. Please note that reserves have nothing to do with this. Please also note how marginal credit worthy customer is very different in boom vs bust. Banks are pro-cyclical, which is why trying to get them to act counter-cyclically reveals profound ignorance.”
As I think I explained earlier, changes to reserve policy have an impact on marginal cost. With respect to IOR, the higher the rate paid on reserves, the higher will be a banks marginal cost of funds because reserves will now compete for banks short-term funding. The higher the reserve rate, the more banks will have to pay for their own funding.
btw, your constant reference to ketchup and ad homs on Scott and others do nothing to strengthen your argument.
9. July 2010 at 05:48
Doc Merlin: Yes, Canada has zero required reserve system. It could have 50% required reserve system. Or 120% required reserve system. It could change required level every day if it wanted, it would not change how bank system functioned so long as Bank of Canada supplied system with reserve when it was net short, and drained them when needed to hit their ON IB rate.
This actually illustrates anon and my point. Reserve is meaningless so long as Fed functions as it is supposed to.
And since you cannot fathom in your brain that bank lending has nothing to do with reserve, you are fixated on fact that somehow some pressure can or should be put on reserve so it “squirt out” in form of loans. I do not know if you are academic in some obscure college, blogging desperately to get name checked by paul krugman, or if you are student drunk on kool-aid by clueless professors parroting from equally clueless greg mankiw textbook.
If Fed wants to lower rate it can just stop paying interest on reserve and FFR drop to zero. If it charges interest on reserve, banks can have negative interest rate on account and force money into cash under mattress. Or safety deposit box. This would drain reserve (Scott Sumner high-fives himself and falls over because his one hand missed other hand) but it does nothing positive for economy.
“No, it doesn’t have to be. It can be withdrawn from the banking system and stored as foreign currency reserves by a foreign government.”
YOu are too funny! This must be your brilliant “system thinking” in action. I hopes to god you are not responsible for any actual system of consequence in real life. If you run nuclear reactor pls tell me what state so I can move far away.
If you have US$ liability, it must be plugged into US Fed, and therefore create US$ reserve. China’s US$ deposit are part of US Fed reserve system just like everyone else.
Maybe you think China has magic machine powered by pixie where $ put in one end and yuan come out other end converted by magic. Given the moronic nonsense on this blog I would not put in past you.
let me tell you how it works in real life. YOu can go to FX counter and try for yourself.
In real life you must SWAP one currency for another. The person who gave you new currency takes your old currency and guess what, needs to put it in bank which accepts deposit in that currency. And it accepts deposit in currency only if it is connected to US reserve system.
Try this yourself. Get some paper Yuan, go to local US bank, and say “I want to put this in my account”. See what they do.
This is very dangerous because it require you to go into REAL bank and see what is does in REAL life. It may mess up your fragile “SYSTEM THINKING”. If they do not accept your yuan, or offer to convert it for dollar, just grab them by lapel, and scream in their ear “WHAT ABOUT PIXIE DAM YOU?!!” Then they will know you are economist and take appropriate measure.
9. July 2010 at 09:09
thruth: It is not my fault that macroeconomic field is one big punch line after another, nor is it my fault that Sumner sets himself up as punching bag by opening mouth when others keep theirs shut (even if they believe the same garbage, they are at least troubled by its obvious stench.)
And you confuse me with someone trying to win popularity contest. I am not two-bit associate professor looking for tenure, or hack dying to be bookmarked on Obama’s blackberry. I just know how this stuff works.
On academic macroeconomist side (and we can include idiot from NY FED who you linked to — what you try to prove? Yes, Fed filled with people who have macroeconomic training and therefore no idea how monetary system works. Hasn’t Fed incompetant been clear to you? Everyone else it is obvious) we have following assertions:
1. A loan is vault cash taken out in suitcase never to be spent again.
2. Chinese have pixie machine that converts $ to Yuan
3. It is credible Fed option to break payment settlement system, so some banks will suddenly find they can no longer clear checks.
4. Fed saying “Such and such bank can no longer clear checks so all banks will lend more” is an “adequate explanation” when it is clearly autistic drivel.
4. When payment settlement system is destroyed, and we have bank runs and credit freeze, people will move money to non-FDIC insured accounts (who are breaking the buck) and banks will start making loans into collapsing economy. This is presented as “solution” without trace of irony. Or pulse. Or cognitive function. Or memory.
5. Having people take deposit out of bank and put it under mattress is “stimulative”. Or anything other than asinine. At least you get that this will reduce reserves, but it seems you have once again lost site of actual economy. That is not fair, it was never in your sights.
6. That reserves (asset) compete with funding (liability) on bank balance sheet. Clearly you do not know what balance sheet is. Reserves do not fund anything.
7. Government pouring profit and capitalization into bank by OIR RAISES their cost of funding. Because getting more profit always your cost of capital in your bizarro world.
I could go on all day. I don’t think reserve requirement level has any impact on marginal CoC. At least mechanism for this is very unclear — and you clearly don’t know what it is given your absolute ignorance on every other facet of this topic.
You are swanning about naked in reality and instead of people saying how lovely your coat is, they are just pointing out you have no clothes. You should withdraw now.
9. July 2010 at 10:40
zanon, you said: “1. A loan is vault cash taken out in suitcase never to be spent again.
2. Chinese have pixie machine that converts $ to Yuan
3. It is credible Fed option to break payment settlement system, so some banks will suddenly find they can no longer clear checks.
4. Fed saying “Such and such bank can no longer clear checks so all banks will lend more” is an “adequate explanation” when it is clearly autistic drivel.
4. When payment settlement system is destroyed, and we have bank runs and credit freeze, people will move money to non-FDIC insured accounts (who are breaking the buck) and banks will start making loans into collapsing economy. This is presented as “solution” without trace of irony. Or pulse. Or cognitive function. Or memory.”
Right now, the banks are awash in reserves, far more than required to cover aggregate deposits. If it turned out that one bank was short reserves, that bank would have to “borrow” reserves from another bank (or sell assets for cash…). Your story is the whole system would collapse and there would be mayhem on the streets. Explain, or just stop debating this stupid point?
I’ll deal with the more substantive part later.
9. July 2010 at 13:23
thruth: Yes, banking system is awash with reserve. It is not expanding credit. Surely that should give you hint there is something wrong with your reserves->lending fantasy.
If one bank in system finds itself short of reserve, then it will need to borrow from another bank yes. This happens all the time. It is called IB market. There is even ON IB market, which is where actual FFR comes from.
Now I am going to engage is SYSTEMS THINKING — are you ready? Maybe if you hit your head with brick first.
If banking SYSTEM is short reserve, then, individual banks which are short reserve CANNOT borrow reserve from other bank. This is because there are not enough reserves to go around at SYSTEM level, and banks cannot create reserve. Only Fed can create reserve.
Such banks will not be able to clear checks and will be thrown out of payment settlement system. This is REGARDLESS of their level of capitalization. They essentially cease to become functional bank.
At this point, Fed can either lend to bank at discount, or decide that bank can go to hell. Bank will start bouncing check and experience run. Other banks will say “oh my god” and hang on to reserve just in case because you can run short of reserve for technical reason that have nothing to do with lending. So, payment settlement seize up, bank run happen, system collapses.
Some bank do not like to borrow IB, or IB ON, and certainly not at discount, so they have various liquidity management and business processes in play to minimize risk. This factor into both operational and capital cost. it is simply one business model out of many, no better, no worse.
But, to understand this point you must undestand difference between individual bank and banking SYSTEM. Academic macro, while it arrogantly boasts it does this, in fact gets it completely wrong. Fallacy of compostion is not its greatest blunder though.
If system is short reserve, IB market will not help and payment mechanism will break down. Only Fed can help.
If system has enough reserve, then IB market works unless, for some reason, an individual bank is isolated and thrown out of IB market (which is possible). At this point Fed either decides problem is liquidty, and lends OR it decides bank has been thrown out because it is undercapitalized and shuts it down. Or, if bank is Goldman Sachs, it simply writes them a big cheque.
9. July 2010 at 17:27
zanon, you said “If banking SYSTEM is short reserve, then, individual banks which are short reserve CANNOT borrow reserve from other bank. This is because there are not enough reserves to go around at SYSTEM level, and banks cannot create reserve. Only Fed can create reserve.
Such banks will not be able to clear checks and will be thrown out of payment settlement system. This is REGARDLESS of their level of capitalization. They essentially cease to become functional bank.”
More likely, they will be assessed a penalty for being short reserves. Then they will be expected to rein in their deposit base. (One way to do that is to start jacking up fees or enticing depositors into non-reserve required investments). If they can’t do that they will fail. All of this is contractionary, which is exactly as any macroeconomist (I’m not one) would argue.
Now, a new example, suppose the system is in a happy equilibrium with just enough reserves. Then the Fed announces that it will be raising reserve requirements incrementally two months from now. Banks will have to start building up reserves by selling assets or reducing deposits. This is contractionary and slightly raises banks’ cost of funds (i.e. more reserves required for a given deposit means more idle capital for each loan made) and makes bank lending slightly less attractive. What is wrong with this story?
(we’ll turn to IOR eventually)
9. July 2010 at 17:30
Zanon, I am glad I bring you “joy.”
anon, You said;
“”If banks are afraid the money will be redeposited in the banking system”
That alone is an absurd statement as a qualification – that money MUST be redeposited in the banking system, by some non-bank customer of a bank. That’s the only way the payee can get value for the bank check/draft or whatever the liability form is.”
No, they can ask for cash.
You said;
“I’ve already explained that the former is a silly scenario, because it would require a doubling of publicly held currency.”
Why is that silly, it happened in Zimbabwe.
I generally agree with the points made by Doc Merlin and Thruth.
Serious question: Are there any post-Keynesian economists at the elite Universities? This isn’t a knock on PK, my views aren’t represented there either. I just wonder who is the “go-to” guy to read.
9. July 2010 at 21:08
thruth: The payment does not clear. I, as human on planet earth, knows what happens when you have money in your account but your check does not clear because the bank you are with is no longer accepted as bank.
Your initial solution to bank run was encourage people to move money into non-FDIC accounts. Your new solutions is to jack up fees to its customers! Yes — both of these are really going to help bank that is having a run on it! Do you even think before you type? Or after you type? Or ever?
Take your example: system is at right level for reserve. Fed announces reserve requirement going up in two months. System only has 1 source of reserves, and that is letting Treasury mature and deposit back into reserve account. If enough of this can happen by deadline all is fine. If not, then system is short reserve and you get bank run.
And you really have trouble with the whole SYSTEM part of SYSTEM thinking don’t you? Or is it THINKING part of SYSTEM THINKING? Or are you just awful at both?
Bank sell asset, that mean someone buy asset. If asset shuffle around in system, reserve is unchange. And if bank reduce deposit where does deposit go? Mattress cash? Do people consume it with ketchup?
“more reserves required for a given deposit means more idle capital for each loan made) and makes bank lending slightly less attractive. What is wrong with this story?”
What is wrong is that reserve do not serve as idle capital for loan. reserve does not serve as ANY kind of capital for loan. That is what you cannot get through your indoctrinated brain.
Scott: There are no PK economists at any elite university. Or even at any second rate university. No one in any academic department of any repute has any clue at all how banking system works. They are all in backwaters, and most of them are 50% cranks. Only an madman would pursue academic career as PK. And if they were not mad to begin with, enduring the idiotic drivel that gush forth from no-nothing academic macroeconomist would surly destroy what sanity they had. Look at the insane ideas Doc Merlin and Thruth babble in this thread!
Your understanding of zimbabwe is equal to understanding of finance, money, or the economy. I am sure that your explanation, something along lines of “Harare central bank should have announced a lowered target NGDP” will be joyous and good for many yucks.
10. July 2010 at 03:57
“I just wonder who is the “go-to” guy to read”
Top 4:
Randall Wray:
http://neweconomicperspectives.blogspot.com/search/label/L.%20Randall%20Wray
Warren Mosler:
http://moslereconomics.com/mandatory-readings/soft-currency-economics/
Bill Mitchell:
http://bilbo.economicoutlook.net/blog/
Scott Fullwiler:
http://neweconomicperspectives.blogspot.com/search/label/Scott%20Fullwiler
10. July 2010 at 04:20
P.S.
Wray’s recent blog stuff is pretty political. He has a lot of deep background writing behind that, including several books.
10. July 2010 at 06:10
zanon, you said “Take your example: system is at right level for reserve. Fed announces reserve requirement going up in two months. System only has 1 source of reserves, and that is letting Treasury mature and deposit back into reserve account. If enough of this can happen by deadline all is fine. If not, then system is short reserve and you get bank run.”
Is a bank run not contractionary? More seriously, as far as I can tell, all you are arguing is that the Fed would do OMOs to undo the contractionary effect of the change in reserve policy. Otherwise, other things equal, raising reserve requirements would raise the fed funds rate from its previous target, which would be a contractionary impact.
“Bank sell asset, that mean someone buy asset. If asset shuffle around in system, reserve is unchange. And if bank reduce deposit where does deposit go? Mattress cash? Do people consume it with ketchup?”
Returning deposits as cash obviously wouldn’t help solve a reserve shortage would it? Mattress cash reduces banking system reserves.
I have two observations. 1. Banks have to hold reserves against a certain subset of total system liabilities, but not ALL liabilities. 2. Outside of the banking system there are additional reserves of cash (the various forms of mattress cash held for whatever precautionary/transactions smoothing need of the holder). So without Fed intervention, I can see two ways to resolve the shortage of reserves in banking system. 1. enticing customers to swap deposits for non-reserve required liabilities by offering better terms on the those liabilities. 2. selling assets in exchange for cash held by the public. To attract that cash, banks would have to reduce the price of assets (increasing their yields). Thus, either way, yields on assets have to go up. The higher yields have a contractionary impact on the economy. With a big enough shortage of reserves, this could be extremely disruptive as you say, but nonetheless, CONTRACTIONARY.
10. July 2010 at 06:56
anon, Thanks for the links. Yes I was familiar with their blog, and had read some of their writings. I will say this for you zenon and Turd. It takes a certain self-confidence to assert that everyone who doesn’t share the beliefs of a couple University of Missouri at Kansas City economists is an idiot. Of course I’m also at a minor league school, but I don’t claim that everyone else in the universe is an idiot.
10. July 2010 at 07:58
Idiot is obviously a strong modifier. But there is no question that the vast majority of the economics profession is ignorant of how central bank reserve accounting works. This is the start of a spiral of misinterpretations about how the monetary system works.
10. July 2010 at 11:35
Kind of glad I took a few days off from this after reading through. By the way, Scott, you’re at a very good school (I’m sure you know that) and I would only hope that I could be as welcoming and pleasant to all comers as you if I ran my own blog, but I know I would fail miserably.
Regarding this:
“Turd, Your example of 13% return on T-bonds is of course irrelevant, as it is the ex ante returns that matter. Ex ante returns on T-securities are very low, even for longer term Treasuries.”
I agree. My point was it turned out to be a good investment and that was to counter your point about “low yielding” Treasuries. As it turns out, though, I went and looked and banks aren’t holding many Treasuries at all, actually, anyway.
“Regarding opportunity cost of reserves: Suppose the Fed raises the IOR to 23%. Would banks rather lend out money to business borrowers, or park the money in safe reserves at 23%. That’s why I think the IOR is the opportunity cost of using reserves for some other purpose. I just can’t see a bank lending money to a business at 8%, when they could lend to the Fed at 23%.”
If you set the IOR to 23%, that effectively becomes the FFR. Lending rates would similarly adjust upward.
I think frequent blog commenter JKH summed up my view quite well a few months back when he wrote the following (link is http://macromarketmusings.blogspot.com/2010/02/feds-exit-strategy.html):
There is no opportunity cost associated with excess reserves provided that excess reserves are earning the risk free rate. This is because excess reserves are zero risk weighted for capital allocation purposes. Accordingly, they require no capital allocation on a risk weighted basis.
It is irrational for any bank risk/capital manager to allocate capital to risk simply on the basis of perceiving a nominal opportunity cost when earning the risk free rate on a risk free asset. It is wrong to view earning a higher risk adjusted return on risky assets as a substitute for lower earning risk free assets, without reference to capital requirements. A higher compensation for risk is a requirement with or without the alternative presence of risk free assets. So the presence of risk free assets is a red herring in terms of opportunity cost.
10. July 2010 at 12:25
Thruth on July 9 @5:46 (and sorry for misspelling your name several times before):
I am assuming that’s the Meulendyke quote (if not, please correct me, as I would be interested in the source), which Basil Moore then ridiculed in his response. At the very least, in order to do what Meulendyke said the Fed could do there, the FFR would go above the target rate, or the target would have to be raised as soon as the Fed closed the window. Also, to do what Meulendyke is suggesting there, the Fed would have to accept substantial volatility in the FFR relative to the target.
10. July 2010 at 12:30
Regarding “go to guys,” among the non-PK, Charles Goodhart (LSE and BoE), Ulrich Bindseil (ECB), and Piti Disyatat (BIS) have argued much the same as PK regarding central bank operations and banks.
10. July 2010 at 19:16
Turd, you said “I am assuming that’s the Meulendyke quote (if not, please correct me, as I would be interested in the source), which Basil Moore then ridiculed in his response. At the very least, in order to do what Meulendyke said the Fed could do there, the FFR would go above the target rate, or the target would have to be raised as soon as the Fed closed the window. Also, to do what Meulendyke is suggesting there, the Fed would have to accept substantial volatility in the FFR relative to the target.”
Yes, it’s the Meulandyke piece and, yes, I agree that would be the implication. (I think I said at some point many posts back that I generally agree with the view that during normal operation the Fed endogenizes reserves to maintain the fed funds rate target)
But, you guys also seem to be claiming something a little stronger: that any change in reserve policy will be fully neutralized by the Fed to maintain the target and that lending/economic activity will be unaffected. Now, it’s not completely obvious to me that the Fed’s provision of reserves to stay at the current target in the wake of the reserves policy change is enough to produce the same level of money/economic activity. A trite example is that the raising of reserves may effectively signal to the market the Fed’s intention to tighten in the future, which will have a contractionary effect on lending today.
As far as reducing the rate of interest on excess reserves, say from .25 to 0, I presume you guys would argue that the Fed would be forced to drain all of the reserves to keep the FFR at .25 resulting in no economic impact? Now suppose the FFR is already 0 and the Fed announces a negative rate on ERs, but doesn’t drain the reserves. What happens next?
A bigger picture question that I have about the view that you guys appear to be espousing (at least in some of the more extreme blog comments) is that a sufficient statistic for monetary policy is the short-term rate. Anything else the Fed does is apparently either meaningless or fiscal policy. By construction the only way out of the liquidity trap is fiscal policy. Is that the right interpretation?
btw, thanks for the macromarketmusings link.
11. July 2010 at 09:48
anon, Indeed, no one outside the University of Missouri at Kansas City seems to understand monetary economics. Not even economists at the University of Missouri main campus. That is pretty sad.
Turd, I still don’t follow your argument. Suppose the interest rate on one year bank loans is 15% and there is a 10% chance of default. So the expected return is 5% on bank loans. Now suppose the IOR rises from 4% to 6%, and let’s call the IOR the risk free rate. I still say that makes banks less willing to make a loan at 15% where the expected return is only 5%.
12. July 2010 at 05:02
“All four campuses have felt their regional designations potentially cause them to be mistaken as second-tier regional institutions. Considerable controversy over use of the generic name has been caused by the Columbia campus claiming de facto ownership, further marginalizing the other three campuses.”
I guess that would be like the New York Fed doing a takeover of the Board of Governors.
That said, I suppose the St. Louis Fed knows squat about implementing open market operations.
12. July 2010 at 06:31
thruth: “a sufficient statistic for monetary policy is the short-term rate. Anything else the Fed does is apparently either meaningless or fiscal policy”
Yes. Although the short-term rate is probably impotent also, as its nets effects on economy are hard to guess, some expansionary and some contractionary.
In bank run, if reserve merely move from one institution to another then it is neither contractionary nor expansionary in monetarist sense. Of course, in bank run lots of other thing happen, so this is ridiculous statement. Not as ridiculous as claim that bank run is now Fed policy tool!
“1. Banks have to hold reserves against a certain subset of total system liabilities, but not ALL liabilities.” Yes. But intra-day they can run overdraft if needed, and ON, if needed, they borrow from each other OR Fed OR payment mechanism breaks. It is caboose.
Look at capital when loan gets written off — much more important to lending which is what we are talking about.
You must teach yourself double entry bookeeping before reading my next para. If you cannot understand this, please become academic economist where ignorance of accounting is requirement and understanding t-table (or reality in any way) will get you denied tenure.
Bank hold loan as receivable (asset). Loan goes bad. Bank recognizes payment stream will not come in, so debits asset. Now, it cannot credit reserve (asset) because it hold reserve as ASSET and nothing has happened to impact reserve! It must debit liability.
So, which liability it debit? Liability in first loss position — equity.
Thus you see how it is equity (liability) which must be set aside to enable new credit, and how equity is limiting factor in quantity of credit bank can create. If bank is short on equity, say, because it knows it has to make massive receivable debits (but has not yet because government makes winks) the QUANAITYT OF RESERVE IS IMMATERIAL to it making new loan.
Sumner: You see? All PK at po-dunk university, and Mosler is not academic is ex-finance who makes ridiculous cars in tropical island and running for office. They are also, i have found, pretty ignorant about micro and policitical economy. Nevertheless, their knowledge of macro, finance, and banking operation is 100% correct and that at top rate university is total 100% garbage. (Please do not misinterpret what I wrote, I do not think Bently is top rate and I also think it is total 100% garbage)
this is state of knowledge in world. may gods help us all
12. July 2010 at 09:51
anon, and Zenon. Fair enough about the campus, but that still begs the question as to why other smart economists think PK is nuts. At least I can claim that top economists have never heard about Sumnernomincs. But PK has been around for many years. I know lots of smart economists who have looked at it and have the same reaction I do.
12. July 2010 at 10:10
I’d be interested to know why they think it’s nuts.
Can you summarize (or Sumnerize) the general nature of the objection?
12. July 2010 at 10:15
P.S.
Do you think people like Krugman and Mankiw are aware of it, but dismiss it as beneath contempt? Or is it so un-mainstream, they are not even aware of it?
I see virtually no references to it outside of PK sites, which suggests the sort of reception you point out.
12. July 2010 at 10:17
P.P.S.
“Sumnerize” has a ring to it.
If you haven’t used it yet, you may freely do so.
12. July 2010 at 12:39
zanon, you said “Yes. Although the short-term rate is probably impotent also, as its nets effects on economy are hard to guess, some expansionary and some contractionary.”
I’m not sure what you mean by this. Are you arguing that because the Fed approximately follows a Taylor rule that rates are endogenous as well? Or something else?
“Thus you see how it is equity (liability) which must be set aside to enable new credit, and how equity is limiting factor in quantity of credit bank can create. If bank is short on equity, say, because it knows it has to make massive receivable debits (but has not yet because government makes winks) the QUANAITYT OF RESERVE IS IMMATERIAL to it making new loan.”
This sounds like you are arguing that the bank becomes capital constrained. Why can’t the bank raise new equity in the wake of losses? If lots of banks were facing losses then cutting FFR would, other things equal, help to lower the cost of new equity, wouldn’t it?
12. July 2010 at 21:00
scott: oh please. you may know nothing about macroeconomics, finance, or the economy, but I am sure you are all to knowledgable about politics in the academy. It should be obvious to you why PKs are totally unrepresented.
Beside. Macroeconomics not only department which is filled with garbage, it just has the worst excuse (in my opinion)
thruth: 1. Something else.
2. Yes, bank can become very much capital constrained, or even fall below capital requirement and be shut down by FDIC.
Bank must raise new equity in wake of significant loss if it can.
FFR is one of many things that impact cost of new equity, and it is never only variable in play. If FFR has an impact to enable additional lending, it is through this CoC channel though. But it is only one of many many factors that impact CoC, and since I do not venerate this one figure I see no point in isolating it or discussing it certus paribus wrt to CoC. Bernanke does not raise or lower FFR just for fun.
Beside, FFR do many things across economy, so even if it lower CoC in one way, it may raise CoC in another. The net effect is unclear and there are also many other confounded factors
13. July 2010 at 08:33
anon, I don’t know enough to answer your question for sure, but years ago when I looked at PK they seemed to argue that monetary policy was endogenous, and hence not the cause of inflation. I thought that was completely wrong. The QTM provides a theory of the price level. Unless I am mistaken, PK (and the economics of Keynes himself) didn’t provide an explanation of the price level, merely the rate of inflation. So the model seemed flawed for that reason.
Perhaps things have changed since, I haven’t read any PK for a while.
zanon, I know even less about office politics–I pay little attention to it.
13. July 2010 at 12:39
scott: Problem with QTM is that economists have no idea what V is. They think bank lend out deposit, or bank lend out reserve, or some such nonsense. Biggest joke is that they keep V constant.
and i apologize for assuming you know something about something. I was not talking about mere office politics, i will not explain what i was talking about as it is in vain.
14. July 2010 at 05:12
Scott: It seems this is one way that PKs would determine the price level
“Full Employment AND Price Stability
There is a very interesting fiscal policy option that is not under consideration, because it may result in a larger budget deficit. The Federal government could offer a job to anyone who applies, at a fixed rate of pay, and let the deficit float. This would result in full employment, by definition. It would also eliminate the need for such legislation as unemployment compensation and a minimum wage.
This new class of government employees, which could be called supplementary, would function as an automatic stabilizer, the way unemployment currently does. A strong economy with rising labor costs would result in supplementary employees leaving their government jobs, as the private sector lures them with higher wages. (The government must allow this to happen, and not increases wages to compete.) This reduction of government expenditures is a contractionary fiscal bias. If the economy slows, and workers are laid off from the private sector, they will immediately assume supplementary government employment. The resulting increase in government expenditures is an expansionary bias. As long as the government does not change the supplementary wage, it becomes the defining factor for the currency- the price around which free market prices in the private sector evolve.
A government using fiat money has pricing power that it may not understand. Once the government levies a tax, the private sector needs the government’s money so it can pay the tax. The conventional understanding that the government must tax so it can get money to spend does not apply to a fiat currency. Because the private sector needs the government’s money to meet its tax obligations, the government can literally name its price for the money it spends. In a market economy it is only necessary to define one price and let the market establish the rest. For this example I am proposing to set the price of the supplementary government workers.”
http://moslereconomics.com/mandatory-readings/soft-currency-economics/
🙂
14. July 2010 at 07:29
thruth, Well they don’t really spell out monetary policy, but there might be a nominal anchor there. Still, it seems like a wacky idea, as it’s hard to know how the system would impact the macroeconomy.
But how is the price level determined in an economy (like ours)that doesn’t have that system in place?
14. July 2010 at 09:21
scott: price level is determined by supply, demand, willingness to pay, budget constraint etc. which all create TRANSACTION at a PRICE, and that price is the… price.
It is very complicated and I do not expect you to understand the connection between prices and individuals exchanging goods and services for agreed upon amount. It actually does live in QMT if you can engage your brain and figure out where the transaction actually exist. I will give you a hint, it does not exist at the Fed.
it is exactly this kind of blind idiocy, and i do mean “idiocy”, that keeps academic macro the fetid backwater of garbage that it so obviously is
btw. I know many “professional” MMTers like employer of last resort as policy, but I do not like it and view it as one manifestation of infantile understanding of political economy and micro you see riddling MMT.
But, what fiat Govt prints and pays for services is one mechanism to determine price level. Maybe the ultimate mechanism in economy that fully embraces reality of MMT. In absence of that, best to focus on more general case (which I describe above)
14. July 2010 at 09:56
Scott, “thruth, Well they don’t really spell out monetary policy, but there might be a nominal anchor there. Still, it seems like a wacky idea, as it’s hard to know how the system would impact the macroeconomy.”
Supposing that policy makers could credibly promise to not tinker with the “supplementary” wage rate (yeah right…), I would expect we would get to a place where the supplementary wage, in real terms, would have less buying power than current unemployment benefits.
🙂
Zanon: aren’t you just defining away the role of the Fed? How does your view differ from Cochrane’s fiscal theory paper? Also, have you seen David Romer’s paper in the JEP “Keynesian Macroeconomics without the
LM Curve”: http://elsa.berkeley.edu/~dromer/papers/JEP_Spring00.pdf
14. July 2010 at 12:57
thruth: I am not “defining” away anything. I am just describing reality. Fed is around, it does stuff, I just point out actual consequence of what it does and why. I do have thoughts on what Fed SHOULD do, but is very different from what Fed is and outside scope of this discussion.
Also, i don’t understand why you confuse me with academic economist or someone who cares about what they say or think. My assumption is that Cochrane and Romer are MORONS who utter garbage. I will not wade through their turgid babble without compensation for time and hazard pay.
15. July 2010 at 09:41
Zanaon, Supply and demand for what? The standard theory is that the S&D for each good determines its relative price, and the absolute price level is determined by the S&D for money.
25. September 2012 at 02:00
I am really flabbergasted by this discussion. Of course, by now I am two years late to the party, and I could not possibly say things better than Turd, anon, and zanon (who seem to be very knowledgable, and zanon really cracked me up…). Nonetheless I can perhaps contribute something for other people who stumble over this discussion (as I just have).
I do not understand the obsession with paying interest on the reserves – it is just another policy tool to keep the interbank lending rate at the Fed’s target rate. Truth has essentially got it right, it is equivalent to doing open-market operations and selling treasuries to drain reserves. It is just another way of stabilizing the interbank rate at the Fed’s target. This is nicely described in a paper of the NY Fed here:
http://www.ny.frb.org/research/epr/08v14n2/0809keis.pdf
So any problem you have is with the Fed’s target rate, which has some small effect on lending at the margin. Whether the Fed achieves the target by open-market operations or by paying interest on reserves does not matter.
I am also befuddled that apparently Sumner and truth seem to see a tradeoff between holding reserves and making loans, i.e. in the sense that the bank obtains some “money” from somewhere and then has to decide whether to hold his as reserves or “extending it as a loan”. This seems to me to be the idea behind the perceived tradeoff.
But there is no tradeoff, since it is not the same kind of money: When banks extend a loan and create a deposit, this is “bank money” that they create themselves at this time.
Reserves that they hold, on the other hand, are *central bank money*, so they can only exist in the account of the bank at the Fed, and they can only be used to settle accounts with either other banks trough the interbank payment mechanism, or to buy something from the Federal government (extinguish a tax liability or buy t-bills).
The only exception is cash, since cash are not bank notes, but “Federal reserve notes”, so if somebody takes out the deposits from his bank account (which up to that point are only “bank money”), it is instantaneously converted into central bank money (Federal reserves). So this does indeed change the reserves that the bank hold at the Fed, since in effect the bank has to withdraw something from its reserve account at the Fed.
But this is the only way that a non-bank member of the public can get their hands on actual reserve money. As long as a deposit is not taken out in cash, it does not affect the banks reserves.
This means that there is no tradeoff between putting “money” into loans, and holding a reserve balance, because a bank simply cannot convert the money in its reserve account into money it pays out in a loan. This can only happen is the loan is in fact taken out in cash – then it becomes reserve notes. But who does that, and to have an effect, we would need a hugely extended cash economy. Not likely.
But all of this has been said before, and probably in a clearer way. “Banks do not loan out reserves (except to other banks).” They can’t, because all “money” in their accounts created by a loan is bank money, not central bank money. Conversion only takes place if people take out cash (“Federal reserve notes”), but banks cannot force people to do this.
So there is no tradeoff between holding reserves and making a loan.
25. September 2012 at 02:18
It should be obvious that there is no tradeoff: The idea that there is some money that can be allocated either for extending a loan or as reserves held is incompatible with what people are saying otherwise.
It seems to be uncontroversial that in extending a loan, a bank creates at the same time a deposit (a liability to itself), and an asset for itself (the loan).
Now what happens to the reserves, and how is this affected by the rate payed on reserves? If the bank already has sufficient reserves to cover the newly created deposit, nothing changes! It now receives interest on the reserves, and it receives interest on the loan as well! So receiving interest on the reserves and receiving interest on the loans happen at the same time! There is no tradeoff!
If it does not have the reserves to cover the newly created deposit, it needs to obtain the reserves on the interbank market, paying the going rate set by the Fed at the target rate. This is going to impact the profitability of the loan, but it has nothing to do with the rate paid on reserves by the Fed. On the contrary: it makes it less costly for the bank to acquire the reserves, since in effect it only has to pay the difference between the target rate and the “floor”. Again, this actually helps to extend the credit, and there is no tradeoff!
For there to be a tradeoff, loans would have to effectively *decrease* reserve balances, but they don’t do that! On the contrary, they *increase* reserve balances, and the Fed made this more profitable!
So where is the tradeoff?