Bank reserves can never leave the balance sheet of the Fed, but that does not limit how they can be spent. Reserves are a form of money and can be spent on anything. However, banks transact with other banks in a different way than how banks transacts with non-banks. This is due to our two-tiered monetary system, where not everyone is eligible to hold reserves. For the cryptocurrency fans: this is the equivalent of a bitcoin holder able to pay another bitcoin address, but unable to send bitcoin to someone with only a Ethereum address. In this note I sketch out a few illustrations that should be helpful in understanding how this works in a bank-to-bank and bank-to-non-bank scenario.
Note: Although banks can spend reserves on anything, bank are heavily regulated in what they can buy. They cannot go out and load up on equities and high yield or other risky assets without violating risk and regulatory limits. Regulation and profitability, not reserves, is what constrains a bank’s balance sheet.
Bank to Bank Transactions
In this example a bank will purchase an asset – be it a Treasury, office building, car etc. from another bank. The transaction is essentially an asset swap, where Bank A swaps $100 in reserves for an $100 asset that Bank B holds. The aggregate balance sheet of the banking system does not change.
Bank to Non-Bank Transaction
In this example Bank A will purchase an asset from a non-bank, who banks with Bank B. Note that this results in the creation of bank deposits. The Non-Bank essentially converts his asset into bank deposits through the sale. Bank deposits are created when a bank creates a loan asset or buys something from a non-bank. Whereas the aggregate level of reserves in the banking system is unchanged, the aggregate balance sheet size of the banking sector is $100 larger. The spending of reserves by any individual bank does not change the aggregate level of reserves in the banking sector, but it does shift the distribution.
In practice, banks must hold a certain level of reserves to meet regulatory liquidity thresholds. For large banks, this is usually to meet the Liquidity Coverage Ratio. The LCR mandates banks to hold a level of high quality liquid assets like reserves, Treasuries, or reverse repo backed by Treasuries in proportion to their expected 30 day outflows. Banks that spend their reserve levels will thus only buy other HQLA assets like Treasuries or Treasury reverse repo. For example, in late 2018 JPM decided to invest a large chunk of its reserves into reverse repo. This was a time when repo rates (proxied by the Secured Overnight Funding Rate) rose comfortably above the Interest on Reserves paid by the Fed on reserves.
Do you know if these types of commercial bank to commercial banks swaps play out much in reality? because If I have a liquid treasury why would I want to swap it for reserves
Or if I have a risky corporate bonds why would the other party want to swap me their safe reserves?
In what scenario would each party both be happy with the swap
When commercials banks swap reserves for Treasuries (reverse repo transaction), they are doing that maximize profit subject to their regulatory constraints. They can earn slightly more on the reverse repo loan than the interest the Fed offers on reserves. Treasuries and reserves are considered equal under liquidity regulations, so it is neutral from a regulatory standpoint.
Commercial banks could swap Treasuries for reserves if they needed liquidity via repo loans. For example, suppose Bank A’s big depositor withdrew a ton of deposits to deposit at Bank B. Bank A will need reserves to meet those withdraws (the reserves will be wired to Bank B). Bank A may repo out their Treasuries to temporarily raise liquidity to meet those outflows. The key here is that inter-bank payments can only be settled in central bank reserves, not Treasuries. In practice commercial bank’s rarely do this today because the level of reserves is so high they don’t need liquidity.
In practice Commercial banks do not lend in reverse repo against risky collateral. This is because their regulatory metrics a lot, and because in the post 2008 crisis world the repo market for risky collateral is much smaller and less liquid. A hedge fund or other investment fund could do this trade to earn a relative high return (the risky collateral will be haircut so risk is not high)
Thanks for the article. The Fed under current law, is forbidden to fund the Treasury. If so how is the massive increase of Fed’s tsy holdings not effectively funding the Treasury? In you example of a bank purchasing an asset from a non-bank, why would a bank purchase a non-interest-generating asset, such as a car? Seems that this is a pathway for excess reserves to enter the private sector directly. Thx!
The law is a tricky thing that can be interpreted in many ways. Looking through history, the U.S. Gov can do many things today that the Founders did not intend. The same for the Fed (why is it fighting climate change?). In practice, it seems the gov does whatever it wants. To your other questions – banks largely hold interest generating assets, but they can also buy things like company cars, office buildings etc. They don’t need reserves to buy them, but can buy them by creating money (bank deposits). Banks don’t lend out reserves, they create money (bank deposits) when they make loans. Reserves are used to for payments to other banks or entities that have Fed accounts.
Thanks. This question come up quite often. When a bank can create deposits in exchange for an asset from a nonbank, what account is that deposit initially created in? Equity? And, to pay the nonbank, reserves (of which there are ample) must come into play to transfer the deposit liability to the nonbank’s bank, correct?
To add on to FedGuy, in practice, banks can buy any asset from a non-bank simply by crediting a deposit account. However, they would be practically limited by buying many things by 1. earning assets. It would eventually not be profitable for the bank by purchasing a lot of company cars as those are not assets which generate cash flow. 2. leverage ratios. This limits overall asset size of balance sheet based on equity, and raising capital from shareholders is expensive if it is just to buy more assets.
When the bank needs to payout / buy Commodities and doesn’t have enough inventory or equipment.
thanks, nice article
“Reserves are a form of money and can be spent on anything.” Where is this information coming from?
From my experience. An easy way to see this is to just look at the Treasury’s TGA account at the Fed. It holds hundreds of billions of reserves that are constantly being spent on transfer payments, missiles, tanks etc.
Is the situation the same for TGA and commercial bank reserves?
Oops should have just written in the same comment instead of 2 different comments. So if bank reserves at the Fed can be used for nearly anything, maybe even buying cars idk, is there a specific list published by the Fed, or US government, specifically stating what banks cannot use reserves to buy? Thanks!
See my comment above. There are natural limits to what banks can buy, mostly from regulation, but I am not aware of any specific list. Banks naturally seek a profit motive within the regulatory guidelines (riskier financial assets have higher capital charges – if they are allowed).
The TGA contains real government funds spendable on anything the Treasury wants, not bank reserves.
It’s more like a checking account than the inter-bank accounts the commercial banks have with the Fed.
The Fed doesn’t buy securities from non-banks at all, it purchases them exclusively from Primary Dealers, all of whom hold accounts at the Fed and are paid in bank reserves.
Also, Commercial Banks can’t buy assets from non-banks with bank reserves, they buy them by creating a new deposit (literally creating new money). Bank reserves only come into the picture to settle any inter-bank balances at the end of each accounting period.
In these ways, bank reserves remain inside the banking system and are not money in the way we usually understand it. The Fed can’t “print” money, only the Commercial Banks can.
But by using reserves to settle non bank’s bank accounting and crediting non bank deposit account for the asset bought, the banks are creating new money out of thin air for buying that asset. That new money is now in economy. But this whole new money credit was only possible due to bank having reserves to exchange for non bank’s bank reserves. In that case aren’t they indirectly using reserves and converting it into more real money for non bank ?
Yes, same comment I had. Ultimately existing reserves will come into play when the spending bank needs to pay the nonbank. I also imagine that shareholder equity comes into play. If a bank purchases an armored car, I assume the shareholders get a piece of its liquidation (after depositors) if the bank goes belly up.
Hi, Fed Guy!
I just came to know you blog recently, and I appreciate your writings very much.
In reading this content, I think there is a more accurate way of describing what’s going on behind scene in case of Bank to Non-Bank Transaction. Here’s what I think more appropriate.
1) When Bank A purchases Non-Bank holding assets, Bank A credit +$100 to Non-Bank’s deposits(DA) and Bank A holds assets of +$100.
2) Non-Bank wants to move its deposits to Bank B, then Non-Bank must draw its deposits from Bank A.
3) Bank A and Bank B do this job via central bank. The end result is as follows. Bank A holds $100 assets, reduced $100 reserves, and also reduced $100 deposits. Non-Bank holds $100 deposits of Bank B. Bank B holds $100 reserves and liabilities of $100 deposits.
So question is whether Bank A purchased $100 assets using reserves or not. I think moving reserves to Bank B is followed by purchasing assets. It seems a little bit confusing to describe the situation as Bank A’s purchasing $100 assets using reserves. Of course, the final result is the same, though.
Thanks for reading!
Frome Korean follower.
As you correctly noted, when a bank purchases an asset from a Non-Bank it pays in deposits it creates out of thin air. The reserves are only needed to settle inter-bank payments should there be outflows. If there is no outflows then no reserves are needed.
The example in the post assumes the Non-Bank does not have an account at Bank A, so it skips your steps 1 and 2 and goes directly to inter-bank payment to Bank B. I think your description is an accurate portrayal of what would happen if the Non-Bank had an account at Bank A, and then moved to Bank B.
Thanks for your quick reply.
It does appear indirectly what makes bank A purchase asset is because they had reserves in the first place. See below
But by using reserves to settle non bank’s bank accounting and crediting non bank deposit account for the asset bought, the banks are creating new money out of thin air for buying that asset. That new money is now in economy. But this whole new money credit was only possible due to bank having reserves to exchange for non bank’s bank reserves. In that case aren’t they indirectly using reserves and converting it into more real money for non bank ?
I look at banks as “liquifying machines”. Takes a non-liquid asset, mortgage, capital equipment, etc, and turns it into liquid money, deposits. So bank-money creation is not so mysterious.
Personally I believe it is a power that should not be given to the private sector bank since the bank preferentially creates money for the wealthier (creditworthy) – has led to the gross inequality we see.
great, clear article. I just found your blog while researching the increase in reverse repo usage and found this. Have also followed on twitter, hope to hear more content from you on recent developments.
Can you comment on what you think the > 1T reverse repo usage is signaling? Or is it just a consequence of QE and the new normal. Do you see this usage returning previous levels? When you say banks are permitted to spend their excess reserves on anything, does that mean they can use it to purchase equities / corporate bonds etc? Is there anything regulatory that precludes them from doing so? By lending to the govt via RRP are they basically saying, we don’t think there are better returns on our excess reserves.
Thanks in advance!
Perhaps one explanation for the high level of reverse repo is the fact that a great big bunch of cash was recently put into circulation by the TGA spend-down. In an era of (transitory) inflation, banks would want to hold an asset that generates a small return, such as treasuries acquired via reverse repo, rather than simply holding cash that is eroding in value. Of course, this leaves the fed “holding the bag” full of weakening cash, but that is a nice way for them to take some of the sting out of inflation for the banks.
I’m sure that Joseph could correct me if I’m wrong about this. Thanks for the original article — it was very thought-provoking!
“For the cryptocurrency fans: this is the equivalent of a bitcoin holder able to pay another bitcoin address, but unable to send bitcoin to someone with only a Ethereum address.”
If, in your analogy, the non-bank has the Ethereum address, what is happening when the (non-bank) bank customer withdraws their deposits in cash?
Hi Joseph,
A bit late to the party here but could you clarify, when banks sell their treasuries to the Fed as part of a QE transaction they get in exchange bank reserves. Can they then use those reserves to go and buy more treasuries. Either at an auction or in the market?
Thanks
Yes they can
Suppose in this scenario the bank buys a treasury at auction using these new reserves as you suggest they are able to do. Does this increase total money supply? No, right? Because this is just swapping labels on the asset side of the bank that bought the treasury and increasing the government liabilities and its general account assets (until it spends that money, which will result in those funds get shuffled among various bank reserve accounts). I get that there is a transfer from the reserve side of the coin into the real-economy side, as bank reserves were seemingly converted into real-economy deficit spending by the government, but doesn’t that added real-economy amount just settle back into the reserve side as folks, say, deposit their government checks? Don’t the banks then take those deposits and send them to their reserve account at the Fed where they hope to flip them into something more profitable like a bond? Where am I missing the net add to total money supply? Or am I correct in thinking there is no increase in this scenario.
Can banks buy ‘trading assets’ with reserves?
I believe the Fed has strict guidelines that banks must adhere to in terms of the risk of the assets they buy.