In fact, if the Fed were primarily concerned with government borrowing costs, it's obvious that the easiest course of action would be to do nothing, let everything collapse into deflation, and watch the world pile into US Treasuries, sending yields even lower than where they are now.Read it at Business Insider
They haven't done this. The Fed is trying to inflate the economy and that means higher rates. The myth that the Fed is motivated by public sector debts is nonsense.
The Biggest Myth About The Federal Reserve
Joe Weisenthal
(h/t Scott Fullwiler via Twitter)
Actually, the myth is this comment:
ReplyDelete"The Fed is trying to inflate the economy and that means higher rates. "
All else equal, yes. But all else is never equal.
It is possible for there to be a prolonged period in which the Fed promises to buy treasury bonds, and bond investors front run the Fed and start to increase the demand for those bonds, which will decrease the yields. The Fed will be compelled to buy the bonds at those higher prices.
This is where we are today. A stagnating real economy due to years of prior inflation, and the Fed promising to buy treasury bonds in up to 10 and even 30 year denominations. The bond market can, for a time, front run the Fed and lower the yields, despite the fact that the Fed is inflating when it buys those treasuries.
Joe Weisenthal doesn't seem to grasp this.
Priceless. Morgan Stanley's Vincent Reinhart is 180 degrees wrong, in about the most clear cut way possible...
ReplyDeleteWow. What a bunch of crazy commenters on that site. Another hangout for zero hedgies?
ReplyDeleteThis comment has been removed by the author.
ReplyDeleteYou know, it really would make things easier if Tsy spent with its own US Notes and the Fed borrowed with its own Fed bonds.
ReplyDeletehttp://traderscrucible.com/2011/11/07/4-ways-to-change-the-fed-a-users-manual/
"Priceless. Morgan Stanley's Vincent Reinhart is 180 degrees wrong"
ReplyDeleteDan, these guys wouldn't be satified with just 178º or 179º wrong.
Hey everyone. A friend of mine recently told me he thought that the Fed purchased bonds directly from the Treasury, and asked me if that is correct. I decided to write up a mini-essay explaining the process (to the extent of my understanding). I'm going to past it below here. Would anyone like to comment on the extent to which I have it correct, and where I might be incorrect? (btw, I know that I left some things out… for example, the fact that as of recent the Fed is payin IOR, etc.)
ReplyDelete…..
When Congress appropriates spending beyond what it collects in taxes, the U.S. Treasury creates bonds. These bonds are basically savings accounts which are held at our Federal Reserve Bank. You probably know them by another name: the U.S. National Debt.
The auction of these bonds occurs between the Treasury and a designated group called Primary Dealers. The primary dealers have a special relationship with the Treasury in that they are required to purchase all bonds that the Treasury is selling. This special relationship guarantees that any time the Treasury auctions bonds, they will always be sold. The primary dealers then sell the bonds in the market.. to people, businesses, institutions, foreign governments, etc.
The Fed’s relationship to Treasury bonds is purely a banking operation...
In the United States, banks are required by law to keep a percentage of all deposits as a “type” of money called Reserves. Reserves are kept in two forms, 1) as vault cash and 2) as a bank account at the Fed called a Reserve Account. These Reserves are used for two functions, 1) to supply money for when depositors show up and want to withdraw cash from their bank account and 2) for inter-bank borrowing and lending.
Meanwhile, the Fed attempts to target what’s called the Federal Funds Rate (FFR). The FFR is the interest rate that banks charge each other when they borrow and lend Reserves from each other in what's called the inter-bank market. Banks borrow and lend Reserves to each other when they are short of or in excess of the percentage of Reserves they are required by law to hold. Usually this percentage that banks are required to hold as Reserves is 10% of deposits. The reason banks don’t want to hold more Reserves than they are required to hold is because usually Reserves earn no interest. That is, they are not profitable for a bank.
Based on simple Supply and Demand economics, the more Reserves there are in the banking system, the more the FFR is driven downward. Likewise, the less Reserves there are in the banking system, the more the FFR is driven upward.
Still with me? :)
(continued…)
ReplyDeleteWhen money is deposited in banks, this increases the amount of Reserves in the banking system. If the Fed were to do nothing, this would push the FFR downward. Likewise, when money is taken out of the system, for example through cash withdraws, this decreases the amount of Reserves in the banking system. If the Fed were to do nothing, this would push the FFR upward.
The way in which the Fed maintains it’s target FFR is through buying and selling Treasury bonds in the banking system, a process otherwise known as Open Market Operations. This process goes on all day every day. If the banking system is short of Reserves (relative to the target FFR), whereby upward pressure is being put on the FFR, the Fed will BUY Treasury bonds in order to inject Reserves into the banking system. If the banking system is in excess of Reserves (relative to the target FFR), whereby downward pressure is being put on the FFR, the Fed will SELL Treasury bonds in order to extract Reserves from the banking system. This is all done to maintain the target Federal Funds Rate (FFR - again, this is the interest rate that banks charge each other when they lend and borrow Reserves from each other).
And that’s it. That’s the Fed’s relationship to Treasury bonds.
Now, recently the Fed has preformed Quantitative Easing I, Quantitative Easing II, and Operation Twist. These procedures are essentially the same process as I just described. The difference between them and Open Market Operations, is that OPOs only target the FFR.
Normally the Fed targets only the FFR, and allows the market through supply and demand to determine the interest paid on bonds throughout the entire Yield Curve… all the way out to 30 year-bonds.. i.e. the “distance” from the FFR to the 30-year bond is called the Yield Curve, and it includes bonds of many different time-lengths. What QE1, QE2, and Op-Twist did is try to flatten, or lower, the interest rate on the part of the curve just past the FFR.. that is, short-term bonds.
Make sense?
I thought that was fairly easy to understand, JK. Good write-up.
ReplyDeleteQuestion. We know that the Fed cannot run out of reserves. But it could run out of bonds if it wants to increase the rate and lower their price. Can the Fed credibly target a higher rate when it is about to run out of bonds?
ReplyDeleteJK,
ReplyDeleteGood job!.. rsp,
Hi Peter P
ReplyDeleteYes. It can (1) raise the rate paid on reserves and its lending rate in tandem to be a corridor around the new rate, or (2) with Govt approval, it could issue time deposits to drain the excess reserve balances (if it sets, say, a 1 week time deposit at its desired new target rate and its higher than the current target, banks will convert ER to time deposits). The Fed has suggested in the past that when it comes time to drain the reserve balances, its "exit strategy" could involve time deposits (not that an exit strategy would ever be necessary, but that's a different point). Both of these are in use in other countries already.
STF,
ReplyDeleteI understand, makes perfect sense. Thank you so much for clarifying!
Great job, JK.
ReplyDeleteI would only recommend one change, since air quotes, speech quotes, scare quotes are all the rage in describing "reality".
You say: In the United States, banks are required by law to keep a percentage of all deposits as a “type” of money called Reserves.
I would "modify" to: In the United States, banks are required by law to keep a percentage of all deposits as a type of money "called" Reserves.
@JK,
ReplyDeleteOverall I think it’s pretty good. A couple items you might consider adding/clarifying (forgive me if I’m being overly picky)…
1) Reserves and vault cash are different creatures. Reserves only live at the FED. When a bank needs vault cash (currency and coin) it buys it from the FED with some of its excess reserves. In aggregate reserves are only good to settle payments between banks, to lend to other banks short of reserves, buy US treasuries, settle US Federal government tax payments, and to buy currency and coin from the government.
2) Why does the FED have a target interest rate? A: The FED is legally responsible through its charter to ensure the safety and health of the inter-bank payment clearing process (the payment system). This means the FED needs to ensure that there are always sufficient reserves available to the banking system to clear and settle payments between banks. How does it know when the system needs more reserves or has too many? It sets a target interest rate tied to a money market, in the US it’s currently tied to the overnight inter-bank lending market known as the Federal Funds Market (target rate is called the Federal Funds Rate).
3) “When money is deposited in banks, this increases the amount of Reserves in the banking system.” I think you’re cutting to the chase too quickly here… Do you want to describe the current federal deficit spending operational arrangement or the theoretical ones? Also, deposits can consume reserves or create reserves depending on what’s happening. If I am a bank receiving a deposit that originated from a loan then excess reserves in the banking system will decrease by an amount equal to the regulatory reserve requirement (assuming one exists). If the deposit is tied to US treasury transaction then reserves are flowing into the system (and depending on how you choose to describe the bond transaction the flow of reserves may be via a direct FED transaction or an indirect FED loan). Depending on the overall demand and supply for reserves the FED will perform OMO to defend its target rate.
4) When you discuss QEX you might want to point out that since reserves are only used for the above items, and banks don’t lend them out, that QE is really just an asset swap designed to change interest rates. In aggregate it’s a portfolio shift of financial assets (interest baring bonds for reserves which earn little or no interest) and NOT “money printing”.
JK,
ReplyDelete"This special relationship guarantees that any time the Treasury auctions bonds, they will always be sold"
Bonds offer risk-free interest.
The alternative for banks is to hold reserves which earn no interest or less interest.
The government can only default voluntarily in its own currency.
Even if inflation is super high, banks would still rather get that interest than not.
"When money is deposited in banks, this increases the amount of Reserves in the banking system"
ReplyDeleteI think you mean when 'cash' is deposited in banks?
Cash held in vaults has no direct effect on the FFR as far as I'm aware. If banks convert cash back into reserves then this could put downward pressure on the FFR.
If in contrast banks expand deposits through making loans, then this could increase the demand for reserves, and if there are no 'excess' reserves available to meet this demand, the FFR would rise unless the Fed did something to stop it (lending reserves, OMOs etc).
"the more Reserves there are in the banking system, the more the FFR is driven downward. Likewise, the less Reserves there are in the banking system, the more the FFR is driven upward."
ReplyDeleteIt depends on the supply and demand, not necessarily just quantity.
This simple model describes in a simplified way the relationship between interest rate and the quantity of reserves.
ReplyDeletehttp://krugman.blogs.nytimes.com/2012/04/02/a-teachable-money-moment/
The demand 'curve' shifts to the right when banks demand more reserves, and shifts to the left when they demand less.
If interest is paid on reserves, the interest remains the same even as the quantity of reserves increases, so the demand curve shifts to the right each time reserves are added.
I think.
* (and the curve shifts to the left each time reserves are removed)
ReplyDelete"(2) with Govt approval, it could issue time deposits to drain the excess reserve balances (if it sets, say, a 1 week time deposit at its desired new target rate and its higher than the current target, banks will convert ER to time deposits)."
ReplyDeleteScott, the Fed does this now,
"The Term Deposit Facility is a program through which the Federal Reserve Banks offer interest-bearing term deposits to eligible institutions."
http://www.frbservices.org/centralbank/term_deposit_facility.html
Thanks for the points of clarification everyone
ReplyDelete@ Adam1
ReplyDeleteto be technical
reserves = reserve balances held at FEd (rb) and vault cash.
Tom, do you know whether quantity of vault cash also has an effect on the interest rate? Is this something that is reported? Dumb question perhaps
ReplyDeletey, vault cash is counted are reserves wrt figuring a bank's required reserves (RR).
ReplyDeleteBanks lend excess rb in the interbank market. When the Fed is not paying IOR and setting the rate above zero, then it uses OMO to adjust excess reserves in the overnight market to hit the target rate.
I understand that bit. I was wondering about something else.
ReplyDeleteDo you know of any links to detailed nuts and bolts descriptions of how the interbank market actually works minute by minute, day to day?
@Tom (and JK)…
ReplyDeleteMy mistake. After re-reading JK’s write up I do see when he first takes about reserves he means reserves in the technical sense – which includes vault cash and reserve deposits at the FED. We spend so much time talking about “reserves” when we specifically mean reserves held at the FED for interbank payment settlements that I started to read, “1) to supply money for when depositors show up and want to withdraw cash from their bank account and 2) for inter-bank borrowing and lending” as a mistake when technically it is correct.
adam1,
ReplyDeleteThanks for the correction.
I ended up responding to the same guy with another write up. He thought the $100….$90….$81… multiplier was how banking worked. Below is my response, feel free to comment on where I'm correct and where I'm not correct…
…..
Banks are in the business of making a profit from the spread of interest payments. It’s only sort-of-accurate to say that banks borrow money from X and lend it to Y, and then the difference in the interest paid to X, and the interest charged to Y, is where a banks profit comes from. There is a counter-intuitive caveat…
Regardless of how many deposits a bank has, banks are able to make loans and “create money” out of thin air. When a bank makes a loan, it effectively creates the deposit at the same exact time. For example, if you borrow $200,000 from a bank, that loan AND the $200,000 deposit into your account happen simultaneously and instantaneously, and that money doesn’t “come from” anywhere. Said another way, the bank does not have to make sure it has enough deposits in its account before it can make that loan. Because it doesn’t have to make sure it has the $200,000 “on hand” … this means that the banks don’t really lend out our deposits.
I need to go sideways, to go forward…
In the aggregate, all banks are required to keep a percentage of their deposits as a kind of money called Reserves. Reserves are held as both 1) vault cash and 2) a bank account at the Fed called a Reserve Account. Most of a bank’s Reserves are held in their Reserve Account. These Reserves are primarily used for borrowing and lending in the inter-bank market, i.e. between banks.
Every bank’s Reserve balance is constantly going over and going under what that bank is required to hold by law. If you withdraw cash or write a check to someone (who has a different bank), this lowers your bank’s Reserves. If you deposit cash or someone (who has a different bank) writes you a check that you deposit, this increases your bank’s Reserves.
When a bank is short of required Reserves, they seek them out… usually by borrowing Reserves from other banks in the inter-bank market.. i.e. from banks that have excess Reserves. And vice versa. Also, banks can ALWAYS get needed Reserves from the Fed, at what’s called the Discount Window, which costs more than acquiring Reserves in the inter-bank market.
So, where do our deposits come into this?
Our deposits are merely the cheapest means by which a bank can increase its Reserves. The Discount Window at the Fed is the most expensive source, the inter-bank market is the second most expensive source, and our deposits are the cheapest source.
(continued)
ReplyDeleteRemember, when you deposit money into a bank account, you are automatically increasing that bank’s Reserves. With the addition of your deposit…
1)) if your bank was short of Reserves, then your deposit just narrowed that gap without your bank having to borrow in the interbank market (again, borrowing in the inter-bank market is always more expensive than the interest your bank will pay you on a deposit)
and
2)) if your bank was not short of Reserves, your deposit adds to it’s excess Reserves, which your bank will then make a profit on by loaning in the inter-bank market to banks that are short of Reserves.
What’s important to highlight is that whether a bank is short of or in excess of Reserves, this does NOT affect its ability to make loans. Banks make loans to credit-worry borrowers regardless of their Reserve position, and then do what is necessary to make sure they have enough Reserves “after the fact” …
This doesn’t mean that banks can loan out an unlimited amount of money. What banks are constrained by is their capital, i.e. assets minus liabilities. The government sets a ratio for which banks can leverage their capital in the loan creation process.
"These Reserves are primarily used for borrowing and lending in the inter-bank market, i.e. between banks"
ReplyDeleteThe primary function of reserves is to settle payments between banks and with the government (inc the central bank). Reserves are currency, whereas bank credit is an IOU for currency.
Say you promise to pay your friend $100, and he later promises to pay you $90. On the day that you settle your debts, you will simply hand him $10 in cash. Or you will transfer $10 from you bank account to his.
Banks are the same, except that they only settle their debts with each other in currency, i.e reserves.
Fundamentally, banks need reserves because they HAVE to settle with the government in reserves/currency. They need them to pay taxes (which are debts imposed by the government) and to repay loans to the central bank. They also need to hold a certain quantity of cash for customer withdrawals.
If you have a $100 account in bank A, and then move your account to bank B, bank B won't necessarily receive $100 in reserves from bank A. The quantity of reserves transferred between the banks is determined by the difference between what each bank owes to the other on the day that they settle. So if bank A owes B $100 million and bank B also owes A $100 million, then no reserves will be transferred between them at all on settlement day.
Capital is basically assets minus liabilities but it's quite complicated as there are different grades of assets that fall into different 'tiers' of regulatory capital...
"When a bank makes a loan, it effectively creates the deposit at the same exact time"
ReplyDeleteMoney creation by banks is just double entry accounting.
When a bank "makes a loan", they simply type into their computer that they owe the borrower the amount in cash (let's say $100).
In practice they create an account for the borrower and credit it with the loan amount. The borrower then has $100 'credit' in his account. This credit is called a 'deposit'.
The credit is simply a promise on behalf of the bank to give the account holder the amount in cash if they ask for it.
In turn the borrower promises to pay the bank the loan amount plus interest over time in either deposit money (bank credit) or cash.
The deposit is the bank's liability, whilst the borrower's promise (his signature on the loan contract) is the bank's asset.
The bank deposit is the borrower's asset, and his promise is his liability (debt).
If you were to take out a $200,000 mortgage, you could in theory ask to withdraw the amount in cash. But what would be the point?
Given that people tend to only withdraw a tiny fraction of their money at any time in cash, banks are able to massively expand the supply of credit, or deposit money, relative to cash/ currency.
This is also why bank credit is called 'inside money', and currency (government money) is called 'outside money'. 'Inside money' only exists inside the banking system, as a record of liabilities and assets, credits and debits.
Good explanation, Anonymous. A ank loan creates an obligation to the bank to repay principal and interest in the unit of account settable in currency, either cash or rb for interbank settlement, e.g, with a check or electronic deposit, although this can be cleared by netting intra-bank or inter-bank.
ReplyDeleteThe deposit is a promise by the bank to deliver currency on demand either in the form of cash or rb for settlement interbank, although this can also be cleared by netting intra-bank or inter-bank.
From this analysis it is clear that the mutual promises are entries on the creditor's and debtor's respective books. The settlement is in principle in currency (cash or rb), which is only created by the govt, netting notwithstanding since all transactions are not settled by netting.
Inside money is essentially mutual promises made inside to settle accounts in outside money, the loan entries and the deposit entries representing the promises that exist inside and currency (state money) coming from outside the vehicle for final settlement after netting.
The fact that a great deal of actual settlement is inside due to intra-bank and inter-bank netting does nor alter this underlying relationship of inside and outside.