Here's a mental model I find helpful for understanding current circumstances:
"Quantitative easing" means issuing new money -- a government obligation that pays no interest -- to purchase treasury (and agency) bonds -- government obligations that pay interest. Until very recently, for good reasons (a global financial crisis, a global pandemic), the Fed and other central banks around the world have been engaged in quantitative easing at an unprecedented scale, replacing government-issued financial instruments that pay interest (bonds) with government-issued financial instruments that pay no interest (money). The result has been an unprecedented increase in private cash balances -- what many call "liquidity sloshing around."
Last year, some central banks started doing the opposite, "quantitative tightening," i.e., selling previously purchased bonds (or letting them mature), removing liquidity (government-issued money) from financial markets and replacing it, directly or indirectly, with financial instruments that pay interest (government/agency-issued bonds). The result has been a gradual decrease in private cash balances -- one could call it "liquidity evaporating."
For example, you can see the value of the financial instruments the Fed owns (i.e., it has purchased them in the past and continues to hold them) here:
PS. I'm talking only about readily observable facts, not about "excess liquidity" in the abstract sense, e.g., as described by economists who call themselves Keynesians.
I'm not sure what you mean by "excess liquidity ends up in assets." Keep in mind that asset prices are set at each instant by the marginal buyer and the marginal seller. If someone buys a single share of, say, TSLA for twice its most recently quoted price, the market cap of TSLA would instantly double (until the next trade is executed). Prices can rise or drop a lot, even if little money trades hands.
If you're asking how the net present values of long-lived assets change as a consequence of quantitative easing, the answer lies in the impact of quantitative easing on long-term interest rates. All else being equal, when long-term interest rates rise, net present values decline; when long-term interest rates decline, net present values increase.[a]
For example, when the Fed engaged in quantitative easing from 2008 to 2022, it did so expressly with the intention of reducing long-term interest rates. Since last year, the Fed has been engaged in quantitative tightening (selling bonds or letting them mature) expressly with the intention of pushing long-term interest rates up.
--
[a] Asset prices (market caps) eventually tend to follow net present values, usually in fits and starts.
Commercial banks actually create the money (by issuing loans) that is used to buy houses (mortgages) and stocks (leverage).
Central banks affect this process by adjusting the rate to which they lend to the commercial banks, and by quantitative easing /tightening which has a similar effect on long term rates.
Rates are low, more loan value is issued (because the income stream servicing the loan translates into a larger loan amount), asset prices go up. And conversely.
The same way previously created money ends up being used for anything in the private sector: by the actions of individuals, businesses, and non-governmental organizations. If money is cheaper to borrow, they may choose to borrow more, or take on more risk, or what have you.
But money does not "go into assets." That's a misconception. Money trades hands: For every buyer of a house there is a corresponding seller, and for every buyer of a share of stock there is a corresponding seller. Asset prices can rise, or fall, with each trade.
> How would you describe the situation when you purchase a treasury bill then?
Your cash (i.e., money) goes to the seller of the treasury bill.
If the seller is a private investor, your cash goes to the private investor (e.g., a mutual fund, a pension plan, an individual).
If the seller is the US Treasury (i.e., you bought a newly issued treasury bill), your cash goes to the US Treasury, which will deposit it, and later on, will use it to pay for the federal government's expenses, including bond interest. (Recall that, unlike the Fed, the Treasury cannot issue newly created money. The Treasury must borrow or collect taxes from the private sector to fund federal spending.)
If the seller is the Fed (through one of its primary dealers, acting as an intermediary), the trade is quantitative tightening.
>> Recall that, unlike the Fed, the Treasury cannot issue newly created money
This is where it begins to unravel and fall apart. What you say is true in theory only, it’s not true in practice:
The fed finances the primary dealer banks that participate in treasuries auctions - it accepts treasuries as collateral for repos.
The primary dealer banks are obligated to stand ready to purchase treasuries and the Federal Reserve ensures there are sufficient reserves to do so by supplying them through temporary repos (a matched purchase of Treasury debt with a requirement that the seller must repurchase later). While the Federal Reserve is not in that case directly buying the new issue directly from the Treasury, it uses the open market purchase to buy an existing bond in order to provide reserves needed for a private bank to buy the new security. The end result is exactly the same as if the central bank had bought directly from the Treasury.
Unless you purchased a treasury bill from the central bank as part of a money-draining operation, the money is still there. Your counterparty has sold a treasury bill and received money for it, that she'll most likely spend elsewhere.
> How does newly created money (which first goes in commercial bank reserves) finally ends being used to buy houses and stocks?
Well, you have some newly created money, whilst the demand for holding money balances (which depends on the price level and the volume of economic activity) stays the same. So what happens is that money is exchanged away like a hot potato until the demand for money balances rises to match the extra created money. In the short run, this is a mixture of higher prices and a higher volume of expected economic activity, both of which would raise asset prices.
>> you have some newly created money, whilst the demand for holding money balances … stays the same
This is a contradiction. You can’t create money without a demand for it first. In this specific case through the demand for money in exchange for treasuries/MBS/etc.
>> So what happens is that money is exchanged away like a hot potato until the demand for money balances rises to match the extra created money
This view derives from monetarist theory, it’d be fair to say this view enjoys less support today than it did in the past. As with all macro views, it’s primarily BS with perhaps a little bit of truth that may or may not apply in any given real world scenario. Probably not a useful model.
> You can’t create money without a demand for it first.
Why not? At the individual level it literally works the same as any purchase of existing assets. In practice, the counterparty of that transaction will probably spend that money in turn on something else that she actually planned to hold.
> This view derives from monetarist theory, it’d be fair to say this view enjoys less support today than it did in the past.
Well, the biggest flaw of monetarist theory is that it treats "the creation of money" as if it was somehow special, whereas what really matters is the product of money and velocity. (Velocity can be seen as a reflection of external changes in the demand for money balances. It also explains how money can seemingly be "created" out of thin air by entities other than the central bank; what we're really seeing in these expanded money measurements is higher velocity for the actual "high-powered" money that the central bank issues.)
> You can’t create money without a demand for it first
>> Why not?
To be clear i’m discounting stimulus checks which would be exactly that but it wouldn’t be right to claim this is a common source of money creation.
The most common source would be A commercial bank issues a loan creating new money, but without a customer demanding a loan, there is no ability to create money.
The second most common source would be the government spends into the economy by consuming on its own behalf - there needs to be something for sale in the economy (inc. labour / public sector employment).
Usually new money is created when the central bank issues it in exchange for government bonds. It's practically always possible to do this. You could posit a theoretical situation where government bonds are literally indistinguishable from money, but that just means that government bonds are money, so the government treasury has merged into the central bank. Then the central bank has to buy something else, which means it incurs some risk. Or the government can issue more government bonds. "Spending money in the economy" is like this, assuming that the government bonds are permanently rolled over, and never bought back in full.
The economy is always inflationary. Money today is worth less than it is tomorrow. Money that the bank has is just rotting away, becoming less valuable over time. Banks need to take the cash they have and invest it in something to offset the inflationary losses. Because bonds weren't paying much interest, it was a better return for the bank to loan out the money for mortgages, investors, etc. Eventually those loans become a part of someone's paycheck or into their bank account (ie home sale that ended up with a 2-300% return). Compared to the bank buying bonds which essentially removes money from the economy.
This is a very valid question. Up until 2018 or so the liquidity was provided to banks which in turn controlled how much money to loan out. Hence keeping the inflation in check and reducing their risks. however, due to COVID our governments directly handed money to general public through various bills and benefits. That liquidity found its way into other assets. The low mortgage rates pushed housing markets to new highs increase by 100%-200% further increasing net worth of millions who in turn are spending more (wealth effect) further fueling inflation. Now how do you tame this spending? Only way to do so is to reduce the wealth effect or accept the inflation to be higher than 2% target and admit it will take several years to stop it from increasing.
The Taylor rule is showing that the interest rates should be over 10%.
Issuing new reserves not new money. New money can then be issued by the counterparties of the Fed’s open market operations The counterparties are the “primary dealer” banks (theres around 30 of them), these are the banks whose reserve accounts at the fed get topped up in exchange for the assets the fed wishes to buy. This is the US model, the UK model is a bit simpler (replace the entire faux market with the BoE’s asset purchase facility or APF).
>> replacing … bonds … with money
this is basically the effect and you did say you were describing a model not necessarily the actual system but i’d be remiss not to point out the model you describe is not faithful to how the system operates
>> The result has been an unprecedented increase in private cash balances
This is a function of both private debt (150% GDP) and public debt (125% GDP) in the US. Private debt in the US is more of less ignored by many economists but we know from history that this is a mistake regardless of the kinds of stories certain macro economists prefer.
>> The result has been a gradual decrease in private cash balances
Too early to say yet. There are signs private credit growth has continued despite increased interest rates, in which case cash balances could be higher.
Reserves are money -- they are the key component of the monetary base, included in all money aggregates.
Yes, the Fed trades with the rest of the world only via its primary dealers. But note that these dealers are non-US-government entities (specifically, they're for-profit businesses, part of the private sector), or trade with the Fed acting as intermediaries for other non-US-government entities (businesses, individuals, etc., also part of the private sector). Thus, newly issued money with which the Fed pays to purchase instruments in open-market transactions ends up in the hands of... non-US-government entities -- mainly domestic businesses (e.g., mutual funds), domestic organizations (e.g., pension plans), domestic individuals (e.g., day traders), etc., all part of the private sector.[a] The newly issued money ends up as private cash balances.
[a] For simplicity, I'm excluding foreigners from this mental model. I'm also excluding the so-called "multiplier effect" of bank lending.
Reserves can be cash, but the reserves issued through quantitive easing are not cash. They are not a form of money that can be spent in the economy.
>> Thus, newly issued money with which the Fed pays to purchase instruments in open-market transactions
Newly issued reserves, not money. The reserves go to the primary dealer banks. The mechanism by which this results in the banks being prepared to issue new money (for spending in the economy) is not direct. Empirically shown through the muted money creation in response to the enormous reserves injections of QE.
The reasons for this lethargy are many but it’s fair to say capital requirments are the main one. Lending operations (the mechanism by which QE is supposed to ultimately inject real money to the economy) are not reserve constrained. Since Basel, they’re capital constrained.
As I mentioned in my comment, the Fed's primary dealers can and do act as intermediaries for the rest of the private sector. If you sell a treasury bond you own in your brokerage account, it could well be the Fed buying it on the other side, through one of its dealers -- and vice versa.
Otherwise, I agree with you that banks are capital constrained in their ability to make loans. But as I mentioned elsewhere on this thread, I left bank lending (and its impact on higher-level monetary aggregates) out of this mental model for simplicity.
> Reserves are money -- they are the key component of the monetary base, included in all money aggregates.
Money exists on a spectrum. Reserves are money-like in some aspects, but so are bonds and equities, as is gold. These are all stores of value and media of exchange.
QE is creating new reserves (which is money-like) and buying other types of assets (which are money-like): basically an asset swap. The balance sheet often stays exactly the same.
> replacing government-issued financial instruments that pay interest (bonds) with government-issued financial instruments that pay no interest (money).
Except that the Fed is not giving "government-issued financial instruments" with QE. They are placing reserve credits in the banks' reserve accounts. Bank reserves cannot be used in the wider economy, but only with-in the Federal Reserve inter-bank settlement system.
Cullen Roche has a good series of articles on quantitative easying:
> The result has been an unprecedented increase in private cash balances -- what many call "liquidity sloshing around."
In essence, QE is/was an asset swap: bonds for reserves. There was zero net change in the balance sheet: $100M of bonds was exchanged for $100M of reserves.
Yes, bonds for reserves, and vice versa. But recall that reserves are money -- they are the key component of all measures of money. Also, recall that the primary dealers which trade with the Fed routinely act as intermediaries on behalf of third parties -- mutual funds, companies, individuals, etc. During QE, the Fed was buying bonds previously held by the private sector. Now, with QT, the Fed is letting the treasury/agency bonds it holds mature (it may be selling some of those bonds too), canceling the money it receives, and thus putting the private sector in the position of having to buy the new treasuries/agency bonds that refinance the recently matured ones. So, I think it's OK to simplify things as the parent comment did for the purpose of having a useful mental model.
Adding to this, the key is what these market actors purchase in place of the bonds that were sold.
They shift out and buy more risky assets. Corporate bonds, equities.
And then next the people that held the corps/equities that were sold buy still more risky assets (e.g., speculative equities, VC) and so on (e.g., crypto).
"Quantitative easing" means issuing new money -- a government obligation that pays no interest -- to purchase treasury (and agency) bonds -- government obligations that pay interest. Until very recently, for good reasons (a global financial crisis, a global pandemic), the Fed and other central banks around the world have been engaged in quantitative easing at an unprecedented scale, replacing government-issued financial instruments that pay interest (bonds) with government-issued financial instruments that pay no interest (money). The result has been an unprecedented increase in private cash balances -- what many call "liquidity sloshing around."
Last year, some central banks started doing the opposite, "quantitative tightening," i.e., selling previously purchased bonds (or letting them mature), removing liquidity (government-issued money) from financial markets and replacing it, directly or indirectly, with financial instruments that pay interest (government/agency-issued bonds). The result has been a gradual decrease in private cash balances -- one could call it "liquidity evaporating."
For example, you can see the value of the financial instruments the Fed owns (i.e., it has purchased them in the past and continues to hold them) here:
https://www.federalreserve.gov/monetarypolicy/bst_recenttren...
--
PS. I'm talking only about readily observable facts, not about "excess liquidity" in the abstract sense, e.g., as described by economists who call themselves Keynesians.