Submarines have always been an underdog's weapon to some extent, ever since they were used by the South in the US civil war. If you expect your navy to be outclassed, but still want to build one, stealth is one of the few options.
Yes, you can, much better in fact. ATP makes sense in the specific context of the inside of cells, but it doesn't compare to e.g. methane + liquid oxygen in terms of cost or performance.
There's also a big middle ground. For example, some semi-green power is generated by breaking wood into flammable gases + carbon ("biochar"), then burying the carbon and burning the gases.
This is true but not very informative. Each greenhouse gas absorbs heat at different wavelengths, with no single gas covering the spectrum. Adding CO2 or methane increases heat absorption, even in air that's saturated with water vapor.
Leaving aside your geographic confusion, you're wrong about how water is used. Silicon Valley has a tiny water demand per dollar of output; they could even afford desalinization if needed. The industry that would die would be farming.
The farming that would die is the very heavily subsidized business of essentially exporting California water. It's not like Californians will starve if we lose our cotton industry or if we don't have alfalfa to export to foreign dairies.
The LA basin supports over 10 million people. Cut its water supply and the missing water has to be made up somewhere. I know full well it LA is 6 hours away from Silicon Valley but make LA run dry on water and CA-wide water restrictions would only get harsher. Remember that Silicon Valley still has a human population to support, too.
Desalination has been the technological breakthrough of next Tuesday for most of my life. I’ll be happy to see an actual deployment. Until then we wait.
Not sure where you're getting your misinformation, but LA uses less water than the farms of the Central Valley or Southern California.
Cut the water supply and LA will do fine, but the farms will die. (Millions of voters vs. a handful of relatively minor donors. It's an easy decision.)
I have taken several trips to Los Angeles without using a car. It's not convenient, but it works if you allow enough time - the bus lines do at least exist.
You are right that no one knows the exact money supply. But banks printing money is a misconception. Private lenders "create" money by fostering trust - I have $N in the bank, even they also loaned 90% * $N to other people. Effectively the same money is counted many times. But they don't have the right to print money the way a central bank does. They can only re-loan money that was deposited by others ("inside money") whereas the central bank creates money truly at will ("outside money").
> they don't have the right to print money the way a central bank does
Yes, they do. A dollar created by JPMorgan Chase out of a loan wires exactly the same as a dollar created by the Federal Reserve out of purchasing a Treasury. (They both look exactly the same in your bank account and withdraw exactly the same as cash.)
But when they have loaned out as much as they can, they will only be able to loan more money if they get more deposits (or capital). [Edit: I failed to mentions other sources of financing, including from central banks as lenders of last resort. RobertoG is right.] They cannot create money at will as a central bank can. [Edit: the point stands, the intervention of central banks may be required to inject liquidity when the banks cannot create enough money by themselves and cannot find anyone willing to lend money to them.]
You are wrong. After creating money by giving loans, a private bank have to get reserves proportional to the money it has loaned. For this, they got to the inter-bank market or the central bank.
The goal of the Central Bank it's to control the interest rate, not the quantity of money. If there are not enough reserves in the inter-bank market, and the Central Bank refuses to supply them, the interest rate will go up.
So, if the Central Bank want to control the interest rate, it can't control the quantity of money and vice versa.
Modern Central Banks target the interest rate, not the quantity of money. The quantity of money in the economy it's automatically regulated by the level of economic activity.
They target the interest rate buy printing money to buy bonds
"The central bank influences interest rates by expanding or contracting the monetary base, which consists of currency in circulation and banks' reserves on deposit at the central bank"
When they target the interest rate through the monetary base, they lost control of the monetary base.
In other words, if they target the interest rate they can't control the monetary base, so, Central Banks don't control the quantity of money in the economy.
If they choose to target the quantity of money in the economy, the tool that they have to use is the interest rate. So, they would lost control over the interest rate.
Modern central banks target the interest rate: they sacrifice control over the monetary base. The central bank don't decide the quantity of money in the economy, they decide the interest rate.
The implication is that they have to cover any necessary reserves by the private banks if they want to keep the interest rate in its target. So, yes, private banks create money when they lend and destroy money when the money is returned. And that's OK because that means that the quantity of money adapts to the activity of the economy.
> They target the interest rate through the monetary base
This is correct. (The Fed has other tools beside open market operations, however, including directly manipulating rates on reserves and the discount rate.) But monetary base != money supply. The Fed manipulates rates to effect changes on the money supply.
No, because the dollar they created and wired away based on the 10 cents of reserve they hold might be itself deposited. Or the 10 cents might be the deposit of a bank-created dollar. It's a feedback system, it doesn't strictly need a central bank to work.
I’m sorry, it seems I misread what you wrote. That doesn’t contradict what I wrote.
Yes, if the money created is in the end kept in the system as a deposit it enables new loans (but it is not necessarily so if the money goes to someone that uses it to cancel a loan of their own, for example).
And due to the fractional reserve requirement the quantity of money cannot be increased without limit by commercial banks. That’s just a mathematical reality, notwithstanding the fact that it’s not usually a binding constraint.
Right, there's an asymptote that caps the maximum amount of money in a system based on some initial M1 and a reserve fraction. But in practice that doesn't get hit because not all money is lent, so per upthread I think it's perfectly reasonable to say that "banks create money" in the abstract.
We agree. But note that this is in the context of the following exchange:
>> they don't have the right to print money the way a central bank does.
> Yes, they do.
Where the response ignores that that limit exists (may not be relevant in practice but it is in theory: a commercial bank can only give the loans that it can support with its existing reserves or those it may have access to, while for a central bank sky is the limit).
Not understanding the point. Money is fungible. If a bank gets $100 from somewhere, it has to deposit $10 to the corresponding regional branch of the Federal Reserve, but so what? There's no tracking of those specific dollars. They can deposit the ones they got that are "known" to be unreserved cash holdings or they can deposit ones that they just got from a wire transfer from a credit account issued by another bank.
That is what is meant by interest rate targetting: the central banks guarantee to supply as much reserve as needed to maintain the interest rate target.
As long as everyone has confidence in Chase, yes, their accounts are money just like any other money. The difference arises when confidence fails - rare, but important for understanding the banking system. In this case, Chase can run out of money, but the Fed cannot, because they create money by different mechanisms. This is the distinction I'm trying to make.
When a bank is lending you money it's creating the money they deposit in your account. This process is totally independent of people saving money in the bank.
The private bank is forced to search for a quantity of reserves proportional to the money it has created, but it does it a posteriori. The credit department only responsibility it's checking if loaning you money is a good business.
This is like saying that human exertion is independent of having to eat, because a person can go for a walk without first planning their next meal. This appears true on a small scale, but in the long run the inputs and outputs must balance.
For really big transactions, it becomes obvious. If I borrow $1 trillion from a bank, and immediately transfer it out, the bank can't get that money a posteriori. It can't transfer more than it has. It needs to borrow the entire sum, plus the reserve cushion, from someone else - other banks, depositors, or the Fed.
To be clear, I am in no way denying that bank lending increases the money supply. The contrast I'm drawing is with the Fed itself, which really can lend $1 trillion without getting it from anyone else.
If you move your loaned $1 trillion to another bank, your bank will have to find the reserves from other banks or from the Fed.
If it find it in other banks, no new reserves have been added, so, certainly, new money is in the system without the central bank doing nothing.
If it don't find the reserves in other banks at a good price, it will have to ask it to the Central Bank.
Now, the Central Bank have two options, it can give the reserves to the bank or it can refuse. If it refuse the private bank have to get the reserves in the inter-banks system. It have to. Otherwise it would be breaking the law.
So, it will be willing to pay anything necesary for the reserves. That will create additional demand in the reserves, and that mean that the interest rate will go up.
If the central bank wants to keep the interest rate in its target range it has to (it has not other option) to give the reserves to the bank.
Ergo, the central bank can choose the quantity of money in the system or the interest rate, but not both at the same time.
Modern central banks target the interest rate. Ergo, all this quantity of money thing is nonsense.
Your analogy breaks down because my eating and walking will not transfer energy to another human.
My own household experience with money and all analogies failed me when I first started thinking about these things. It only clicked once I started drawing up balance sheets.
As for the giant loan and transfer, I think that might be cleared soon-ish and only settled, with an interbank or fed loan if needed, at a later point in time after netting with other transfers.
Banks do create money by lending. You should be able to educate yourself on this by a quick Google of the question "do banks create money". If you want to dive deeper in this fascinating topic, I can recommend Khan Academy: https://www.khanacademy.org/economics-finance-domain/core-fi...
Lenders do increase the money supply, and in that sense create money, but I am hesitant about that phrasing because it elides the important distinction between ordinary banks and the central bank. That's why I mentioned the distinction between "inside money" and "outside money." In the spirit of your comment, I invite you to Google these terms!
I am unclear on a point. Federal spending goes through bonds process to provide the funds, and creates a 'debt'...i.e. the national debt, which some people are obsessed over. When a bank lends money to a private party, this money is created also, ultimately through the same bonds process, I presume? If so, do bank loans increase the 'national debt'?
The national debt refers only to the debt of the federal government. Separately, statistics are also kept on the private debt within a country - in the US, that's about $27 trillion. For comparison, the US national debt is $23 trillion. People care more about the national debt because it is, in a sense, owed by everyone.
Nations, companies, states, towns, nonprofits, and all kinds of other groups can issue bonds, which are then bought and traded on markets in pretty much the same way. There's nothing special about being able to issue bonds - the market is regulated, but the main requirement is simply that people expect you to pay them back.
The biggest difference between bond issuers is that greater trust leads to lower interest rates and thus cheaper borrowing. For example, the US government sells bonds with lower interest rates than any company - buyers demand more return for a corporate bond, because a corporation is more likely to go bust.
I am asking about the whether or not thr money created by bank loan and federal fiscal spending is ultimately created through the same process at the Federal Reserve.
I think your confusion comes from thinking there is only one type of money.
For simplification, there are two kid of moneys, the 'government money' and the 'lending money' (my terminology).
Government money would be: cash and reserves.
Reserves come from two sources (both are the government) the Central Bank and the treasure. So, when the treasure spend in the economy is creating government money. Also, when the central bank 'give' reserves to the private banks is creating money. When the govenrment tax is destroying money and when the Central Bank retire reserves of the system is destroying money.
The 'lending money' is created by the private banks when lending, but it disappears when the money is payed back. This is a different kind of money, a less 'real' money if you want.
You can find more information in many places. For instance:
Say I gave him an IOU for a dollar. When he goes to the supermarket, they will accept the IOU. So will the gas station, his landlord or anybody else he does business with. Their banks will also accept my IOUs. They have all accepted my IOUs before. They all have, or know people who have, converted my IOUs to hard cash before.
The IOU is an account balance. Have I created money?
> Say I gave him an IOU for a dollar...Have I created money?
Technically speaking, yes! (Neat, huh?)
If the IOU is accepted as currency, that is. You have rights to the original dollar and the borrower has rights to the IOU. Practically speaking, however, you haven't created money because nobody treats your IOU as money.
(Side note: this isn't a stupid question and doesn't deserve to be downvoted.)
Only if his obligation to you is transferable - that is, if you can freely pass it onto another person, and then the debt is owed to that other person. For example, he could write you an IOU for one dollar, and then you could give that dollar IOU to another friend (possibly in exchange for goods and services). In that case yes, you have created money.
I think it is more along the lines of:
Say you have no money.
Your friend has no money too, and needs some.
You lend him a dollar (because you are a bank and have special powers).
The superpower of banks is to take the dollar from you and lend 90 cents to your friend while promising that when you need your dollar they will have enough money around to make you whole. (When the time comes, they may or may not be able to keep their promise.)
I'm sorry you didn't like that simplification. Anyway, it's better than "you have no money but you can make it appear from nowhere and lend it because you are a bank and have special powers".
The more details are given, the less interesting the special powers of commercial banks seem to be: not only is the amount of money that they can create limited to some multiple of the "real money" base, but they don't even lend that much out because it's not profitable.
But the "real money base" will expand on demand if the banks need it. It's not a limiting factor! The loan happens first. The base money is created (if needed) after the fact. It's not as if base money is pushed out there and then the base money is used to create loans. Loans pull base money into being.
It’s not a limiting factor anyway because US banks have 1.4 trillion dollars in excess reserves deposited at the Fed (down from 2.7 trillions five years ago!).
The monetary base (physical currency plus banks deposits at the Fed) is currently $3.3tn.
M1 (circulating cash plus on demand deposits) is $3.9tn (the money multiplier is slightly above 1, but it was below 1 for most of this decade!).
M2 (including also savings accounts) is $14.9tn. M2/MB is 4.5, it has ranged between 3 and 5 over the last 10 years. It was between 8 and 12 in the four decades before 2008.
That special power is literally the distinction between bank and central bank, and what this whole thread is about.
Central banks do have that special power, they can literally create money out of nothing. A normal bank can lend out a $, but they need to have taken a $ deposit, and it stays on the banks balance sheet.
Normal banks special powers are turning risky illiquid debt and making it appear risk free and liquid to depositors.
No, because you don’t have the right to use $1 at any time.
In a bank, Person A gives the bank a dollar, the bank loans out (let’s say the full dollar for simplicity) and, some steps later, the dollar is out back into the bank (let’s say person B does this for simplicity).
Both A and B have title to a dollar, so yes, there’s more than one.
The bank created it by creating a liability for themselves. This is ok because long term the loaning of the dollar should create more than a dollar in value, so the liability will be covered.
So you don’t think the bank is creating money when it lends out the balance from savings accounts? (That is, the deposits that you do not have “the right to use at any time”.)
> you don’t think the bank is creating money when it lends out the balance from savings accounts?
Banks don't lend out balances. "Banks lend out balances" is just an analogy. (It was truer under the gold standard.)
Banks create money through lending and destroy money via write-offs. Periodically (in the U.S., at the end of the day) banks count up what they owe and what they're owed and make sure the ratio is legal. If not, they borrow (from each other or from the Fed) to make up the difference.
(For monetary purposes, savings accounts are checking accounts. Banks don't segregate savings from their balance sheets. Balances in savings account can be withdrawn just as easily as those in checking accounts.)
> And for monetary purposes, savings accounts are checking accounts
There are different definitions of money and saving accounts are included in M2 (“money and close substitutes”) but not in M1.
What about time deposits? For short maturities they are included in M3. And what about longer term depositss? Are banks creating money when they lend those out?
Loans create deposits. Whether those deposits then end up in a checking account through a payroll service or a vendor's savings account through a payment for goods and then transfer is irrelevant.
A bank making a loan is trading one asset, cash, for another asset, a loan. If that's "creating money", then doesn't that imply that a loan is "money" and cash is not "money"?
I’m not familiar with any savings accounts that dont permit you to take money out whenever. You might mean CDs, which will, as per the agreement, assess a fee against you if you take it out early. But you still have title to the cash whenever you want. It’s yours. And if a fee happens, the fee is still money that is now owned by the bank instead.
Down the line you suggested depositing money creates money because now you and the bank have a dollar. That’s not true, because the bank doesn’t have title to the dollar.
If I ask you to hold my drink, you’re not allowed to drink it. Banks have rights to do certain things with your deposits, but it’s not theirs. They can’t decide to pay salaries and deduct the amount from the deposits.
It is part of M0. It's more "money" than deposits (M1) or savings accounts (M2). In a discussion about money supply it definitely counts as money. And they have it. Yes, they need to keep enough reserves but apart from that it's their money to use as they please (lend, invest, spend, pay dividends...).
The deposit is their liability (obviously they can't spend it). The dollar is their asset. When you deposit the money into a bank that dollar is not yours anymore, the bank is not keeping it in custody for you.
Banks don't lend any specific deposits. The deposit is a liability of the bank. The deposited cash is an asset. When the bank makes a new loan, they create both a new asset (the loan) and a liability (the deposit of the funds lent). The existing savings account isn't part of the new loan.
While this is technically true, it is a partial description and a bit misleading.
When a loan is taken, it is usually not taken just to sit in a bank account, it is to be spent. When that happens (by cash withdraw or transfer to a different bank), a bank must release/transfer base money, which it can get from cash deposits, incoming transfers or in exchange for some other assets.
Therefore, although banks create M1 money by lending, they are limited in practice by necessity to keep balance of base money. So it is more a distributed/emergent behavior, where each bank ability to create M1 money is limited by its market share and the rate of money production by other banks.
You’re assuming the loan stays at the bank as a deposit, but normally one takes a loan to give the money to someone else. As soon as the client goes to the teller to get her money, or a cheque she wrote is cashed, or whatever, the balance sheet of the bank will change again (and shrink unless whoever gets the money deposits it again at the same bank).The asset (the loan, the fact that the client owes money to the bank) will stay. The liability (the money the bank owed to that client) will disappear, together with some of the liquidity that the bank had in its balance sheet. It’s ability to lend money has disminished and eventually it will need more liquidity and/or financing (in the form of new deposits, interbank loans, bond issuing, capital raising, central bank credit lines or whatever).
Yes, there are of course limits to bank lending. Ultimately, the limiting factor is the ability of banks to make profitable loans. The capital and the reserves will follow if the profitable loans are there.
The money lent does not go away when the check or whatever is cashed at another bank. The reserves follows the check. We're just working with more balance sheets now.
As long as the loan asset remains in the books, it earns interest which will add to bank equity.
We can keep bringing up details, but the basics remain the same. Bank lending creates money.
We agree, money (under various definitions) is created within limits.
In fact banks also create money when they take cash deposits (and destroy money when they hand out cash).
If I deposit a $100 bill into a bank we’re increasing the money supply because I have $100 and so does the bank (until it spends it, trades it for a non-money asset or lends it out).
Very true! I'm just trying to explain that there is a distinction between central banks and commercial banks, objecting to the statement `The central bank "prints" a lot [of] money, but so do private banks.`
I think that the difference is important, because if one believes that a commercial bank can print fiat money the way a central bank does, then the idea of a bank run or bank collapse looks incomprehensible.
I'm surprised and humbled that a project of this importance doesn't have their own physical ARM machine for benchmarking. Compare that to the computer and human time saved across all the places it could be used!
This work was sponsored by ARM itself, though, so it is a bit surprising they didn't supply suitable hardware from one of their licensees (they must have some to do benchmarks themselves).
SolidRun has a MiniITX board supporting up to 2x32GB for around $1000 [0]. They also have a preproduction units for another board for $550 supporting as much RAM [1].
They probably had their reasons, but I'm just showing what's available.
Well cross-compiling (aka "cross translation) is made complicated because PyPy doesn't use a normal toolchain, but that's not the problem here. The PyPy benchmark suite itself takes that much memory to run.
Don't want to be rude (nor rehash old topics) but is pypy used that much to be considered important? People with performance goals go to C, whether its a library or some Cython.
Don't get me wrong, i love pypy and the rpython jit.
