5 ms·

Can you explain why it is needed to grow and shrink the money supply ? I don't get that. In my view, the only problem with gold was that it was not practical to

by popol12 10y ago

Can you explain why it is needed to grow and shrink the money supply ? I don't get that. In my view, the only problem with gold was that it was not practical to use. That's where bitcoin is better, you can divide a single unit into much much smaller unit (up to 10^-8 Bitcoin for instance) so everybody can use it.

erebus_rex 10y ago
Most central banks in the world do a lot of open market operations to keep the economy in check. It is how they affect inflation, interest and exchange rates. Most cryptos are still commodity currency; the commodity is computation. Because we don't decide how many coins get "minted" our ability to affect the money supply is reduced. So we're castrating one of our major interventional instruments. Practicality isn't the issue. During the Gold standard, the bars were kept in vaults and we traded in convertible paper notes. The mint could always print notes of smaller or larger denominations if needed, but couldn't print more than it had bars to convert to. And that is the problem.
- EdHominem 10y ago
  > couldn't print more than it had bars to convert to. And that is the problem. To be fair, one man's problem is another man's solution. "It can all be taken away" isn't a selling point to the currency's users, only to the issuer. Monetary policy has always been a tool of governments, but that's been because the people haven't had any practical options other than another government's currency. Having a flexible layer of fiat lets us have a shock absorber between the markets and the commodities we need to live. But too much of a decoupling and you run into more problems the other way. It's hurt a lot of people when their government hyper-inflates the currency, or steals much of it back via bank controls, etc.
- nickik 10y ago
  I don't see a well working monetary system as a 'interventional instrument'. Its a inherent property of the system, not a outside system doing an intervention. Only since the time of central banks was this considered a 'intervention', because in many ways the central bank is the outsider. > During the Gold standard, the bars were kept in vaults and we traded in convertible paper notes. The mint could always print notes of smaller or larger denominations if needed, but couldn't print more than it had bars to convert to. And that is the problem. Maybe I am reading you wrongly, but the a ability to print larger or smaller notes is totally not relevant. A bank or minting house would actually always print more nodes/coins then they had bars to convert to gold. That is why minting coins was profitable in the first place. With Bank notes this became much more sophisticated. The amount of notes in circulation was not directly determinate by the amount of gold, but rather by the flow of demands for gold (usually by other banks). As a bank you can get away with having 100x 1$ bills in circulation but you only had having 2$ worth of gold in your vault as long as the avg demand for gold per day is only 1$ worth. The other 98$ are invested in interest paying assets. Thus a bank always tries (tried) to maximize the amount of notes in circulation, but were always limited because other banks would demand gold if they got hold of your notes. This system will thus lead to a situation where a increase in velocity (or decrease in demand) will automatically lower the amount of money in circulation because otherwise banks would not be able to hand out enough gold. This system effectively keeps nominal spending stable and that almost exactly what central banks are trying to do (its in fact better then inflation targeting if you look at most models).
nickik 10y ago
There are different ways to explain it, I will use the simplest one. This formula essentially says that goods get exchanged for money in the monetary economy: M(oney)*V(elocity)=P(price)y(output) Lets say you destroy 50% of M overnight. This will have a huge effect on Py, because prices can not change that quickly, you will have a falling output. Now its pretty easy to keep M constant (as long as politicians can not fund their pet projects with printing). The problem is that Velocity (another word for money demand) is not at all stable. So if something happens that for some reason effect money demand, it will have a effect on Py, that is price or output. In the short term, prices can not change, thus output suffers. Every good monetary system has a way to balance this out. Note issuing private banks (before central banks) would automatically do this because when the noticed increasing demand, they would have to raise the gold reserve ratio. If the notice a decrease in demand, they could decrease the gold reserve ration and thus make more profit. A modern day central bank will say that they target inflation. So they attempt to make P stable (predictable) and let the market (supply side) determine the output. If there is a huge crash in V the central bank should be able to see this and before it affects P or y to much, adjust it back. They are making it quite hard on themselves. Many people now call for targeting NGDP, that just means MV. Bitcoin is fine as long all your doing is making payments, once you have multi layered complex economies that depend on the correct working of the prices system to aggregate information threw the system you will have constant problems. Price systems are the secret sauce that make markets actually work as well.
nhaehnle 10y ago
Look at this graph of weekly M2 data, and zoom into any short time period so that you can actually make out individual weeks: https://fred.stlouisfed.org/series/WM2NS https://fred.stlouisfed.org/series/WM2NS Note the seemingly odd monthly pattern that is apparent everywhere along the graph? That's because the economy has a distinct monthly rhythm of loans being used and repaid e.g. for monthly payroll payments etc. You may think that this could be managed with a fixed money supply, but why should it? The days where money was an object are long gone (if they ever really existed). Money is instead a reflection of contractual relations in the economy. As those contractual relations change, so does the money supply. None of these changes in the money supply have to do with the government or the central bank, by the way.[1] The whole fluctuation is due to commercial banks creating money when they issue loans, and destroying money when the loans are paid back. It's a system that works well. A sibling comment mentions MV = PQ (or MV = PY), which is a good point, although they get all confused about the direction of causality (as is common). For example: when the price level increases, the loans taken out by economic actors become bigger, which increases M. Similarly, correlated growth of M and Q is usually really driven by a hidden variable: for example, an increased volume in consumer loans will drive up both Q and M at the same time. Similarly for an increase in investment. Conversely, in a recession, a shrinkage of Q may go hand in hand with a decrease in V (economic actors keep their money for longer rather than spending it) or M (economic actors try to deleverage). [1] They are related, actually, but only because the government is an economic actor. There is no special relation between government and broad money supply aggregates in today's monetary system.
- nickik 10y ago
  I think you are confused about MV=Py. You can never reason from a price change without knowing the reason for the price change. 'When he price level increase', how? why? We know that money demand is non constant. That just has to do with the flow of money. We know that base money is constant unless the central bank act. Thus causation goes from changing V or central bank action to changes in price or output.
  - nhaehnle 10y ago
    It's true that why the price levels change can make a difference. What I wrote was meant as examples of the type of relations that are there. You're still wrong or at least extremely misleading in your characterization of how the money supply behaves. First of all, base money is really almost entirely an implementation detail of interbank settlement and decoupled from the real economy.[1] Because of that, the focus on base money is wrong-headed and you should look at broad money instead. Broad money grows and shrinks almost entirely due to commercial bank lending and loan repayment. The central bank has no influence on that. It is technically correct that base money can only be changed by central bank actions, but that is a somewhat misleading statement, because the focus should be on what the chains of causality are. We've had a perfect experiment for that in the last few years due to quantitative easing. There, central banks tried to affect the real economy by being the initial cause of a change in the money supply. Well, they successfully raised the size of base money to totally unprecedented levels, but that had virtually no effect on the broader money levels, let alone on the real economy. All other central bank actions that change the size of base money are actually done as a reaction to demand by commercial banks, which in turn is a consequence of commercial bank lending. Because in any case, central banks haven't even usually targeted the size of the money supply as a matter of policy. The policy has been to target a level of interbank interest rate, and let the size of the base money supply go wherever it needed to in order to accommodate the interest rate. And when central banks tried to target the money supply via quantitative easing, which was only tried when the interest rate was already effectively at 0, the evidence is that it had virtually no effect. There is no causal chain in practice[2] that starts with the central bank deciding to change the size of the money supply and goes into changes in price or output. [1] It's even decoupled from the major financial markets that people are usually interested in. [2] Of course, in theory central banks could just go nuts and start targeting the money supply in crazy ways. But even in that case, they can't just decide to cut or grow the money supply arbitrarily (because they need to find a commercial bank counterparty for what they want to do). And further, due to the decoupling of base money and broader money, the causal chain would go mostly via the effect that the base money changes have on interest rates, which is a very different story from the one that you probably wanted to tell.

by popol12 10y ago

erebus_rex 10y ago

Most central banks in the world do a lot of open market operations to keep the economy in check. It is how they affect inflation, interest and exchange rates. Most cryptos are still commodity currency; the commodity is computation. Because we don't decide how many coins get "minted" our ability to affect the money supply is reduced. So we're castrating one of our major interventional instruments. Practicality isn't the issue. During the Gold standard, the bars were kept in vaults and we traded in convertible paper notes. The mint could always print notes of smaller or larger denominations if needed, but couldn't print more than it had bars to convert to. And that is the problem.

EdHominem 10y ago
> couldn't print more than it had bars to convert to. And that is the problem. To be fair, one man's problem is another man's solution. "It can all be taken away" isn't a selling point to the currency's users, only to the issuer. Monetary policy has always been a tool of governments, but that's been because the people haven't had any practical options other than another government's currency. Having a flexible layer of fiat lets us have a shock absorber between the markets and the commodities we need to live. But too much of a decoupling and you run into more problems the other way. It's hurt a lot of people when their government hyper-inflates the currency, or steals much of it back via bank controls, etc.
nickik 10y ago
I don't see a well working monetary system as a 'interventional instrument'. Its a inherent property of the system, not a outside system doing an intervention. Only since the time of central banks was this considered a 'intervention', because in many ways the central bank is the outsider. > During the Gold standard, the bars were kept in vaults and we traded in convertible paper notes. The mint could always print notes of smaller or larger denominations if needed, but couldn't print more than it had bars to convert to. And that is the problem. Maybe I am reading you wrongly, but the a ability to print larger or smaller notes is totally not relevant. A bank or minting house would actually always print more nodes/coins then they had bars to convert to gold. That is why minting coins was profitable in the first place. With Bank notes this became much more sophisticated. The amount of notes in circulation was not directly determinate by the amount of gold, but rather by the flow of demands for gold (usually by other banks). As a bank you can get away with having 100x 1$ bills in circulation but you only had having 2$ worth of gold in your vault as long as the avg demand for gold per day is only 1$ worth. The other 98$ are invested in interest paying assets. Thus a bank always tries (tried) to maximize the amount of notes in circulation, but were always limited because other banks would demand gold if they got hold of your notes. This system will thus lead to a situation where a increase in velocity (or decrease in demand) will automatically lower the amount of money in circulation because otherwise banks would not be able to hand out enough gold. This system effectively keeps nominal spending stable and that almost exactly what central banks are trying to do (its in fact better then inflation targeting if you look at most models).

nickik 10y ago

There are different ways to explain it, I will use the simplest one. This formula essentially says that goods get exchanged for money in the monetary economy: M(oney)*V(elocity)=P(price)y(output) Lets say you destroy 50% of M overnight. This will have a huge effect on Py, because prices can not change that quickly, you will have a falling output. Now its pretty easy to keep M constant (as long as politicians can not fund their pet projects with printing). The problem is that Velocity (another word for money demand) is not at all stable. So if something happens that for some reason effect money demand, it will have a effect on Py, that is price or output. In the short term, prices can not change, thus output suffers. Every good monetary system has a way to balance this out. Note issuing private banks (before central banks) would automatically do this because when the noticed increasing demand, they would have to raise the gold reserve ratio. If the notice a decrease in demand, they could decrease the gold reserve ration and thus make more profit. A modern day central bank will say that they target inflation. So they attempt to make P stable (predictable) and let the market (supply side) determine the output. If there is a huge crash in V the central bank should be able to see this and before it affects P or y to much, adjust it back. They are making it quite hard on themselves. Many people now call for targeting NGDP, that just means MV. Bitcoin is fine as long all your doing is making payments, once you have multi layered complex economies that depend on the correct working of the prices system to aggregate information threw the system you will have constant problems. Price systems are the secret sauce that make markets actually work as well.

nhaehnle 10y ago

Look at this graph of weekly M2 data, and zoom into any short time period so that you can actually make out individual weeks: https://fred.stlouisfed.org/series/WM2NS https://fred.stlouisfed.org/series/WM2NS Note the seemingly odd monthly pattern that is apparent everywhere along the graph? That's because the economy has a distinct monthly rhythm of loans being used and repaid e.g. for monthly payroll payments etc. You may think that this could be managed with a fixed money supply, but why should it? The days where money was an object are long gone (if they ever really existed). Money is instead a reflection of contractual relations in the economy. As those contractual relations change, so does the money supply. None of these changes in the money supply have to do with the government or the central bank, by the way.[1] The whole fluctuation is due to commercial banks creating money when they issue loans, and destroying money when the loans are paid back. It's a system that works well. A sibling comment mentions MV = PQ (or MV = PY), which is a good point, although they get all confused about the direction of causality (as is common). For example: when the price level increases, the loans taken out by economic actors become bigger, which increases M. Similarly, correlated growth of M and Q is usually really driven by a hidden variable: for example, an increased volume in consumer loans will drive up both Q and M at the same time. Similarly for an increase in investment. Conversely, in a recession, a shrinkage of Q may go hand in hand with a decrease in V (economic actors keep their money for longer rather than spending it) or M (economic actors try to deleverage). [1] They are related, actually, but only because the government is an economic actor. There is no special relation between government and broad money supply aggregates in today's monetary system.

nickik 10y ago
I think you are confused about MV=Py. You can never reason from a price change without knowing the reason for the price change. 'When he price level increase', how? why? We know that money demand is non constant. That just has to do with the flow of money. We know that base money is constant unless the central bank act. Thus causation goes from changing V or central bank action to changes in price or output.
- nhaehnle 10y ago
  It's true that why the price levels change can make a difference. What I wrote was meant as examples of the type of relations that are there. You're still wrong or at least extremely misleading in your characterization of how the money supply behaves. First of all, base money is really almost entirely an implementation detail of interbank settlement and decoupled from the real economy.[1] Because of that, the focus on base money is wrong-headed and you should look at broad money instead. Broad money grows and shrinks almost entirely due to commercial bank lending and loan repayment. The central bank has no influence on that. It is technically correct that base money can only be changed by central bank actions, but that is a somewhat misleading statement, because the focus should be on what the chains of causality are. We've had a perfect experiment for that in the last few years due to quantitative easing. There, central banks tried to affect the real economy by being the initial cause of a change in the money supply. Well, they successfully raised the size of base money to totally unprecedented levels, but that had virtually no effect on the broader money levels, let alone on the real economy. All other central bank actions that change the size of base money are actually done as a reaction to demand by commercial banks, which in turn is a consequence of commercial bank lending. Because in any case, central banks haven't even usually targeted the size of the money supply as a matter of policy. The policy has been to target a level of interbank interest rate, and let the size of the base money supply go wherever it needed to in order to accommodate the interest rate. And when central banks tried to target the money supply via quantitative easing, which was only tried when the interest rate was already effectively at 0, the evidence is that it had virtually no effect. There is no causal chain in practice[2] that starts with the central bank deciding to change the size of the money supply and goes into changes in price or output. [1] It's even decoupled from the major financial markets that people are usually interested in. [2] Of course, in theory central banks could just go nuts and start targeting the money supply in crazy ways. But even in that case, they can't just decide to cut or grow the money supply arbitrarily (because they need to find a commercial bank counterparty for what they want to do). And further, due to the decoupling of base money and broader money, the causal chain would go mostly via the effect that the base money changes have on interest rates, which is a very different story from the one that you probably wanted to tell.