5 ms·

You're right, in a way. But the main issue I'm still struggling with is credit and lending, which are core features of fractional reserve banking. And the more

You're right, in a way. But the main issue I'm still struggling with is credit and lending, which are core features of fractional reserve banking. And the more I think about it, the less the "fractional reserve" model makes sense when applied to bitcoin. If memory serves, this is how banking works today (beware, there are gross simplifications in here): 1. Customer A deposits $1 in Bank A 2. Customer B asks Bank A for $9 3. Bank A now needs to borrow $9. It can legally do this because it can prove to the government that its reserves ($1) are greater than 10% of its outstanding liabilities ($9). It first tries other banks, but if all else fails it can borrow from the Fed. 4. Bank A gives $9 to Customer B. I cannot see how step 3 happens in "bitcoin banking". The whole point is that there's no "central point of trust" and hence no lender of last resort. Or to restate, one of fractional reserve banking's key purposes is to influence the money supply. But this is clearly impossible under a "pure bitcoin" system, where the money supply is fixed by fiat.

maxerickson 13y ago
Step 3 doesn't have to happen for money supply expansion to take place. If the bank lends out $0.90 of the $1, people in the economy will be acting like $1.90 of money exists.
- AnIrishDuck 13y ago
  Right, hence what I said (which is somewhat inaccurate) about a "100% reserve requirement". The bank could lend any amount up to its total deposits, but not above. This is very different than how banking works today. Banks (in the US) can lend up to 1000% of their current deposits. This is a core feature of the fractional reserve system, and is clearly impossible under a pure bitcoin alternative.
  - maxerickson 13y ago
    In practice, they don't. You can compare loans and leases with deposits here (around 75% right now...): http://www2.fdic.gov/SDI/SOB/ http://www2.fdic.gov/SDI/SOB/ I guess you are saying they do accounting shenanigans with the Federal Reserve to make the deposits larger than they really are? (I will admit, I am more interested in aggregate practice here than I am in what the more extreme banks are doing)
    - AnIrishDuck 13y ago
      I appreciate the link. I haven't taken a macroecon or full-fledged finance class, so take my analysis with a grain of salt. If I recall correctly, "demand deposits" are the subject of reserve requirements, not total deposits. The vast majority of deposits are subject to strict regulation and can't be withdrawn on demand (hence the "demand deposit" distinction). They are, in essence, another form of credit. Not a cash equivalent. If we compare demand deposits (1.1T) to securities and net loans and leases (9.63T), we get pretty close to the 10x money multiplier. Because demand deposits are the only things that need to be backed by cash, the rest of the money is essentially re-invested from another institution in the system. Someone didn't come in to the bank and put cash on the table for those deposits. Instead, they told their bank to exchange their credit with another bank. In other words, it's just a change in the ledger sheet between two parties. Simplistically referred to as "created money". The Fed lets banks "create" this money as long as they meet certain rules. The most important being reserve requirements. Getting back to comparison with bitcoin, my original statement ("1000% of deposits") was perhaps an oversimplification. But it's hard to say exactly where bitcoin fits in to a "new banking order". Would it be considered money, e.g. cash equivalent? What then is used as the medium for credit exchange e.g. the other 90% of financial activity? Or is it the unit of account for EVERYTHING, including credit exchange? If so, then how do parties loan money? Won't 90% of bitcoin sit idle for very long periods of time? And does it really make sense to use a deflationary unit of account for an inherently inflationary activity (lending)?
      - maxerickson 13y ago
        I have no idea why you would ignore $8 trillion in interest bearing deposit accounts. To the extent these funds come from other banks, they will show up on the balance sheets of those banks (as loans). Maybe try working from the assumption that what I am saying is (more in the direction of) correct. The banking system may be set up to work in favor of the establishment and bankers, but it isn't a giant fantasy.
        AnIrishDuck 13y ago
        > I have no idea why you would ignore $8 trillion in interest bearing deposit accounts. To the extent these funds come from other banks, they will show up on the balance sheets of those banks (as loans). What? I'm not ignoring those deposits, just pointing out that those deposits are "created money". They are made by banks, not issued by the Fed. The only restriction on their creation is the Fed interest rate on one side and the reserve requirements on the other. In technical terms, they are part of M2, not M0 or M1 [1]. And banks are definitely allowed create M2, though obviously not M0 and M1. > Maybe try working from the assumption that what I am saying is (more in the direction of) correct. The banking system may be set up to work in favor of the establishment and bankers, but it isn't a giant fantasy. What exactly are you asserting? Nowhere did I state that the banking establishment is some kind of giant fantasy. I just made the fairly uncontroversial assertion that banks are allowed to issue more credit than they have cash on hand. In a sense, when they do so they are "creating" money (though this "created" money is technically referred to as M2). 1. http://en.wikipedia.org/wiki/Money_supply http://en.wikipedia.org/wiki/Money_supply
- moron4hire 13y ago
  Just to be clear, fractional reserve banking is the bank lending out $10 of the $1. "10% reserves" means they keep all of the deposited money. $1 deposit leads to $11 in "money".
  - SkyMarshal 13y ago
    That's not what it means. A 10% Required Reserve Ratio means the bank can lend $9 for every $10 it has on deposit. Eg, it must keep $1 (10%) in reserves. However, when that effect is multiplied throughout the entire banking system [1], the system as a whole amplifies the base amount of money in the system. A 10% Required Reserve Ratio equates to a maximum potential money multiplier of 10, meaning the system as a whole can turn a $10 deposit into up to $100 in circulation. [1]: http://en.wikipedia.org/wiki/Fractional_reserve_banking#Example_of_deposit_multiplication http://en.wikipedia.org/wiki/Fractional_reserve_banking#Exam...
moron4hire 13y ago
With fractional reserve banking, it's not that the bank has to keep 1% of the money they have on deposit, it's that the bank gets to invent 1000% of the money it has on deposit. And it doesn't have to borrow money from the Fed in order to do it, it gets to make up the money on its own. Steps 2 and 4 should read $10 and Step 3 doesn't exist. And step 4 should really read "Bank A marks Customer B as having $10". The vast majority of the money in our economy is numbers on ledger sheets. From there, Customer B uses it to pay off a debt of some kind with Customer C, who then pays Customer D, etc., until supposedly the money eventually gets back to Customer B in some way. Except Customer C deposited the money in Bank B, so Bank B can now lend out $100 to Customer E. That's why pyramid schemes are illegal, the banks hate competition. So yes, it especially doesn't work for bitcoins, because how do you make a blockchain for fractional bitcoins based on a mined bitcoin? The mined bitcoin is the one that has the blockchain. You can't make a bitcoin transaction without first checking the blockchain for double-spending, say nothing about 10x spending.
- AnIrishDuck 13y ago
  > With fractional reserve banking, it's not that the bank has to keep 1% of the money they have on deposit, it's that the bank gets to invent 1000% of the money it has on deposit. And it doesn't have to borrow money from the Fed in order to do it, it gets to make up the money on its own. Steps 2 and 4 should read $10 and Step 3 doesn't exist. I haven't studied this subject in some time, so can't remember the exact mechanics. This summary sounds accurate. I would like to say that "it gets to make up the money on its own", while correct from one perspective, is another way of saying "other banks can respect its credit (how much money it says it has) if it plays by a certain set of rules". I don't remember if this mechanically occurs via some kind of direct lending or via the Fed sanctioning "Bank A's" credit as satisfying reserve requirements for a "Bank B". The ultimate result is the same. > That's why pyramid schemes are illegal, the banks hate competition. I'm assuming that this is said tongue-in-cheek, but pyramid schemes are an example of perversion of credit and lending on a scale that separates them from banks. In particular, the core feature of pyramid schemes is a rate of return that is unsustainable, and thus requires rapid creation of new customers instead of gradual growth of new wealth. If fractional reserve banking is indeed really terrible, what is the alternative? I think a massive body of research is required for someone to assert the superiority of such a system. > So yes, it especially doesn't work for bitcoins, because how do you make a blockchain for fractional bitcoins based on a mined bitcoin? The mined bitcoin is the one that has the blockchain. You can't make a bitcoin transaction without first checking the blockchain for double-spending, say nothing about 10x spending. This was ultimately my point. The principles behind fractional reserve banking and bitcoin seem fundamentally incompatible. Given that fractional reserve banking has worked for many centuries, any alternatives that are radically different should be approached with caution.
nhaehnle 13y ago
> Or to restate, one of fractional reserve banking's key purposes is to influence the money supply. No. There were experiments with trying to influence the money supply in the past, but it doesn't work. Today, the money supply develops endogenously, that is, it is determined by potential borrowers' willingness to borrow and by banks' determination of the credit-worthiness of said potential borrowers. The central bank is tasked with making sure that the interbank clearing system works smoothly and with setting the short-term interest rate. Also, your story about how banking works today is a bit muddled. Whether Bank A can borrow those 9$ has nothing to do with how many reserves they have, but with quality of the bank's assets relative to its capital. The volume of reserves really does not have a noticeable effect on the behaviour of the banking system, mostly because, as I wrote above, it is endogenous anyway.

maxerickson 13y ago

Step 3 doesn't have to happen for money supply expansion to take place. If the bank lends out $0.90 of the $1, people in the economy will be acting like $1.90 of money exists.

AnIrishDuck 13y ago
Right, hence what I said (which is somewhat inaccurate) about a "100% reserve requirement". The bank could lend any amount up to its total deposits, but not above. This is very different than how banking works today. Banks (in the US) can lend up to 1000% of their current deposits. This is a core feature of the fractional reserve system, and is clearly impossible under a pure bitcoin alternative.
- maxerickson 13y ago
  In practice, they don't. You can compare loans and leases with deposits here (around 75% right now...): http://www2.fdic.gov/SDI/SOB/ http://www2.fdic.gov/SDI/SOB/ I guess you are saying they do accounting shenanigans with the Federal Reserve to make the deposits larger than they really are? (I will admit, I am more interested in aggregate practice here than I am in what the more extreme banks are doing)
  - AnIrishDuck 13y ago
    I appreciate the link. I haven't taken a macroecon or full-fledged finance class, so take my analysis with a grain of salt. If I recall correctly, "demand deposits" are the subject of reserve requirements, not total deposits. The vast majority of deposits are subject to strict regulation and can't be withdrawn on demand (hence the "demand deposit" distinction). They are, in essence, another form of credit. Not a cash equivalent. If we compare demand deposits (1.1T) to securities and net loans and leases (9.63T), we get pretty close to the 10x money multiplier. Because demand deposits are the only things that need to be backed by cash, the rest of the money is essentially re-invested from another institution in the system. Someone didn't come in to the bank and put cash on the table for those deposits. Instead, they told their bank to exchange their credit with another bank. In other words, it's just a change in the ledger sheet between two parties. Simplistically referred to as "created money". The Fed lets banks "create" this money as long as they meet certain rules. The most important being reserve requirements. Getting back to comparison with bitcoin, my original statement ("1000% of deposits") was perhaps an oversimplification. But it's hard to say exactly where bitcoin fits in to a "new banking order". Would it be considered money, e.g. cash equivalent? What then is used as the medium for credit exchange e.g. the other 90% of financial activity? Or is it the unit of account for EVERYTHING, including credit exchange? If so, then how do parties loan money? Won't 90% of bitcoin sit idle for very long periods of time? And does it really make sense to use a deflationary unit of account for an inherently inflationary activity (lending)?
    - maxerickson 13y ago
      I have no idea why you would ignore $8 trillion in interest bearing deposit accounts. To the extent these funds come from other banks, they will show up on the balance sheets of those banks (as loans). Maybe try working from the assumption that what I am saying is (more in the direction of) correct. The banking system may be set up to work in favor of the establishment and bankers, but it isn't a giant fantasy.
      - AnIrishDuck 13y ago
        > I have no idea why you would ignore $8 trillion in interest bearing deposit accounts. To the extent these funds come from other banks, they will show up on the balance sheets of those banks (as loans). What? I'm not ignoring those deposits, just pointing out that those deposits are "created money". They are made by banks, not issued by the Fed. The only restriction on their creation is the Fed interest rate on one side and the reserve requirements on the other. In technical terms, they are part of M2, not M0 or M1 [1]. And banks are definitely allowed create M2, though obviously not M0 and M1. > Maybe try working from the assumption that what I am saying is (more in the direction of) correct. The banking system may be set up to work in favor of the establishment and bankers, but it isn't a giant fantasy. What exactly are you asserting? Nowhere did I state that the banking establishment is some kind of giant fantasy. I just made the fairly uncontroversial assertion that banks are allowed to issue more credit than they have cash on hand. In a sense, when they do so they are "creating" money (though this "created" money is technically referred to as M2). 1. http://en.wikipedia.org/wiki/Money_supply http://en.wikipedia.org/wiki/Money_supply
moron4hire 13y ago
Just to be clear, fractional reserve banking is the bank lending out $10 of the $1. "10% reserves" means they keep all of the deposited money. $1 deposit leads to $11 in "money".
- SkyMarshal 13y ago
  That's not what it means. A 10% Required Reserve Ratio means the bank can lend $9 for every $10 it has on deposit. Eg, it must keep $1 (10%) in reserves. However, when that effect is multiplied throughout the entire banking system [1], the system as a whole amplifies the base amount of money in the system. A 10% Required Reserve Ratio equates to a maximum potential money multiplier of 10, meaning the system as a whole can turn a $10 deposit into up to $100 in circulation. [1]: http://en.wikipedia.org/wiki/Fractional_reserve_banking#Example_of_deposit_multiplication http://en.wikipedia.org/wiki/Fractional_reserve_banking#Exam...

moron4hire 13y ago

With fractional reserve banking, it's not that the bank has to keep 1% of the money they have on deposit, it's that the bank gets to invent 1000% of the money it has on deposit. And it doesn't have to borrow money from the Fed in order to do it, it gets to make up the money on its own. Steps 2 and 4 should read $10 and Step 3 doesn't exist. And step 4 should really read "Bank A marks Customer B as having $10". The vast majority of the money in our economy is numbers on ledger sheets. From there, Customer B uses it to pay off a debt of some kind with Customer C, who then pays Customer D, etc., until supposedly the money eventually gets back to Customer B in some way. Except Customer C deposited the money in Bank B, so Bank B can now lend out $100 to Customer E. That's why pyramid schemes are illegal, the banks hate competition. So yes, it especially doesn't work for bitcoins, because how do you make a blockchain for fractional bitcoins based on a mined bitcoin? The mined bitcoin is the one that has the blockchain. You can't make a bitcoin transaction without first checking the blockchain for double-spending, say nothing about 10x spending.

AnIrishDuck 13y ago
> With fractional reserve banking, it's not that the bank has to keep 1% of the money they have on deposit, it's that the bank gets to invent 1000% of the money it has on deposit. And it doesn't have to borrow money from the Fed in order to do it, it gets to make up the money on its own. Steps 2 and 4 should read $10 and Step 3 doesn't exist. I haven't studied this subject in some time, so can't remember the exact mechanics. This summary sounds accurate. I would like to say that "it gets to make up the money on its own", while correct from one perspective, is another way of saying "other banks can respect its credit (how much money it says it has) if it plays by a certain set of rules". I don't remember if this mechanically occurs via some kind of direct lending or via the Fed sanctioning "Bank A's" credit as satisfying reserve requirements for a "Bank B". The ultimate result is the same. > That's why pyramid schemes are illegal, the banks hate competition. I'm assuming that this is said tongue-in-cheek, but pyramid schemes are an example of perversion of credit and lending on a scale that separates them from banks. In particular, the core feature of pyramid schemes is a rate of return that is unsustainable, and thus requires rapid creation of new customers instead of gradual growth of new wealth. If fractional reserve banking is indeed really terrible, what is the alternative? I think a massive body of research is required for someone to assert the superiority of such a system. > So yes, it especially doesn't work for bitcoins, because how do you make a blockchain for fractional bitcoins based on a mined bitcoin? The mined bitcoin is the one that has the blockchain. You can't make a bitcoin transaction without first checking the blockchain for double-spending, say nothing about 10x spending. This was ultimately my point. The principles behind fractional reserve banking and bitcoin seem fundamentally incompatible. Given that fractional reserve banking has worked for many centuries, any alternatives that are radically different should be approached with caution.

nhaehnle 13y ago

> Or to restate, one of fractional reserve banking's key purposes is to influence the money supply. No. There were experiments with trying to influence the money supply in the past, but it doesn't work. Today, the money supply develops endogenously, that is, it is determined by potential borrowers' willingness to borrow and by banks' determination of the credit-worthiness of said potential borrowers. The central bank is tasked with making sure that the interbank clearing system works smoothly and with setting the short-term interest rate. Also, your story about how banking works today is a bit muddled. Whether Bank A can borrow those 9$ has nothing to do with how many reserves they have, but with quality of the bank's assets relative to its capital. The volume of reserves really does not have a noticeable effect on the behaviour of the banking system, mostly because, as I wrote above, it is endogenous anyway.