I'm not sure about all of this, but one feature of the "cashless society" that needs a lot of thought and addressing is missing. And that's the fact that a government can simply cut anyone off from the payment system if it feels like it (as in Canada last year).
A government doesn't even have to resort to overt violence if they can push a button and take away all of a protester's money.
This non-economist is always puzzled by the terms "cash" and "currency." Their meanings are derived from context. Last year I paid cash for a new pickup truck. I wrote a check, because I had enough cash in my checking account to cover it. At the gas station, I use a credit card. But to get the cash price, I would have needed physical currency. But "currency" is also vague and context-dependent. I hear there are foreign currency exchanges in which all the transactions are made digitally. When I was in the Congo/Zaire in the 1980s, the local currency was all nearly worthless paper. "Hard currency" referred to paper copies of US dollars. So I'm not sure what exactly is meant by a "cashless" or "cash free" society. In my current life, I can think of very few occasions where physical money is required or optimal. Perhaps a quarter for the neighborhood kids' lemonade? Or a $20 bill for the school fundraiser, only because the kid begging for it doesn't have a card reader? I did have a fellow drive to my house from about 120 miles away to buy my old pickup, and he had $16,000 in $100 bills for payment. But that's the kind of transaction, oddly enough, that will be flagged by the feds, because he must have withdrawn it from a bank account. (I dutifully reported the transaction to the state Dept. of Licensing, and put a copy in my tax file, just in case.) If they were counterfeit, he could have gone to prison for doing essentially what the treasury/fed does every microsecond. I'm sure there are day laborers being paid every day in physical money. But going cashless and having electronic payment transactions "on record" rather than transferring stacks of special pieces of paper printed with numbers would probably be better for all concerned.
I doubt your starting assumption that a doubling of the money supply will cause a doubling of prices; to me, that sounds like the "neutrality theory of money" disputed by *both* Hayek and Keynes (and many others).
Some people (such as those that sell things to the government) will notice the new money before others, so some prices will rise before others.
Your conclusions may still be valid, but its not persuasive to base them on a contested premise.
"Imagine that we had never relied on cash in order to trade. Instead suppose that we had always used computerized ledgers as the payments system." This is a really fun thought experiment. Filling in some important details. I assume that 1) there is a common unit of measure so that all accounts on all ledgers are denominated in the same "unit". Lets call it a unit. 2) The initial number of units had to be positive and is fixed. I have a strong intuition that the value of assets (measured in units) would be determined by the quanitity of units and according to your story, the price of things would as well.
"The main point that I hope you will consider is the possibility that in a cash-optional society access to the payment system is not controlled by raising or lowering the quantity of any particular asset that you can define as money."
I think this is going to take more explanation as I'm really struggling to see what's new here. I grew up in a small enough town that businesses had 'counter checks' from local banks that people would fill out so they didn't have to carry cash *or* physical checks, and I remember them disappearing as people got credit and ATM cards. Does anybody outside very small towns actually do old school charge accounts anymore? I can understand your illustration of wanting to have traveler's checks (I used to do that even when traveling in the US back in the day) versus leaving the country with nothing but a credit/ATM card (I've done the same thing several times in recent years). I just don't see this as something unique as much as it represents a simply a change in the form of access to the payment system, and especially that the payment system simply casts a wider net than it used to. An ATM swipe now transfers the money in your bank account the same way signing over a traveler's check did back in the day. There just wasn't an internet to perform the transfer back in 1970. I've been working since the 1980s and I've never received a paycheck other than by direct deposit.
I just don't think there's a good case to be made that the 'money supply' has ever been restricted to simply what the government calls currency in a fractional reserve banking system. When was the last time a bank gave a business physical cash as a loan? I don't even think that would have been common since the 1800s or even earlier.
I guess my point is that access to the payment system has never been restricted by currency supply except in the most extreme circumstances. I'm betting people who lived in company towns didn't have to carry much cash either but that wasn't a result of Fed policy.
Having liquid assets in tumultuous times is important and it is wise pointing out access to payment systems. This access is where real censorship and/or control can occur. This is why I believe there is value in having an allocation to Bitcoin. It is volatile, but it’s fixed supply should give it advantages over the long run in highly inflationary periods. It is also importantly a payment system. With the lightning network attached to Bitcoin, you can do instant peer to peer payments for a fraction of a penny across any distance. It will be interesting to see how the dynamics Arnold describes here interact with independent payment rails competing with traditional payment systems. Capital controls will be very tough in this new paradigm. Good luck everyone.
Wait a minute. Our current cash-optional system has cash and electronic equivalents, and they inter-convert through ATMs and banks. Most of the time it does not matter which you use for a transaction, unless you are doing something (like visiting a massage parlor) where you don't want a record kept of the transaction, or if you are using a business that refuses to take cash.
The price of cars, electronics, washers (& still allowed dryers rather than only sun), clothes & toys; plus both food and oil/energy, will continue to be priced more by competition & fighting over market share. Doubling income doesn't double the amount quantity of shoes bought (maybe Imelda?) nor the amount of food bought, tho it might shift quality preferences. But calling a $200 pair of shoes "twice as much shoes" as a $100 pair of shoes does not accurately capture the differences.
The non-asset inflation of increasing prices will level off, at a much higher price level, but then be more stable while we see, again, financial & house asset inflation. So that the rich get richer faster than the workers.
This occurs with or without cash.
Very interesting ideas about the differences going cash-less, but the much MUCH greater ability of gov't to control folk should make all freedom lovers oppose it.
"When the pandemic hit, output of goods and services went down, because people were staying home. But the government sent out checks to everyone, and those checks increased perceived wealth. If perceived wealth goes up and output goes down, the difference is bound to show up sooner or later in prices."
I get that the pandemic hit on output (and related supply chain disruptions) was temporary even if in hindsight it was a little less so than many thought but I have a hard time understanding why Powell, Yellen, Krugman, and other supposedly knowing economists thought the increased spending, and resulting inflation, was short-term, transitory or otherwise not a continuing factor.
.
"If wealth were to suddenly jump by 10 percent, then after some period of time (say, ten years), spending will have been 10 percent higher than it would have been otherwise. Assuming that the actual path of goods and services produced is the same as what it would have been without the increase in perceived wealth, 10 percent more spending will translate into 10 percent higher prices than would have been experienced otherwise."
It took me a minute figure this one out. At least I think I did. It's an odd hypothetical. Wealth jumps 10% without a corresponding increase in production. It can happen but it seems more likely to confuse than clarify the piece. Do I have this right? … If prices increase 10%, was there really a 10% increase in wealth? … Which leads me to believe secondary effects note mentioned might be important here.
.
"As you know, I think that going forward the government needs to use inflation as the only means to deal with its mountain of debt. I expect that the Fed will have no power to stop this inflation."
I don't understand how this "deals with" debt other than maybe short term. Inflation causes interest on the debt to increase as much or more (eventually more). I'm not aware of any country with huge debt that came out of it without either a sacrifice to pay it down or a default.
.
"If the government doubles the money supply,"
This kind of misses an important part of what happens. It seems your whole piece does. While govt can add or remove money from the economy in a direct way, it also does it indirectly by changing interest rates it controls. That influences how much money banks and others will lend. Lending leverages the govt created money supply many times over.
.
"So that’s the best I can do to explain dollar prices."
I don't understand how "your best" says nothing about the velocity of money. That seems rather important to me.
"If we had always lived in a cash-optional or cashless society, then classical monetary theory would never have occurred to any economist."
As you say elsewhere, we currently live in a cash optional society. I don't get why this is different from it always having been this way. I don't get why it means we wouldn't operate the same under CMT if we'd always been cash optional, much less economists never would have thought of CMT. What am I missing?
Similarly, I don't get why using ledgers instead of cash makes any difference unless we also somehow remove money from the ledger system. My credit card and bank accounts are ledger systems that offer greater convenience for most but not all transactions but in the end it seems to me they do the exact same primary function as cash.
"In cash-optional society, access to the payment system is not determined by the quantity of money."
This seems true but I'm missing the point. It seems true for a system that is entirely cash based and most that are entirely cashless too. Is that incorrect?
"The main point that I hope you will consider is the possibility that in a cash-optional society access to the payment system is not controlled by raising or lowering the quantity of any particular asset that you can define as money."
Or maybe I have totally missed what you mean by cash optional. Do you mean dollar (currency) optional?
It seems like access to the payment system is not a binary value. Instead, my access to the payment system is denominated in currency (even if I never hold currency).
I have a checking account with an ATM card attached to it. I can spend all of that, which counts as some number of dollars of access to the payment system. Presumably, the ratio of dollars in that account and access to the payment system is something like 100%. (I’m having trouble disentangling access from willingness to spend. For example, is the point where my balance falls low enough that I have to pay a monthly account fee relevant?)
I have a credit card with a limit. The credit limit seems to represent dollars of access to the payment system. Again, that seems like 100% access for the amount under my limit. In fact, I can probably exceed my limit, but that’s not 100% possible. Maybe there’s a formula there, like my bank would let me go 5% over before just shutting me off.
Then we get to sources of access that are clearly less than 100%.
I have financial assets that I can quickly convert to use for payments. Assume exchange traded funds outside a retirement account. I’d end up having to pay taxes in any capital gains, and there would be brokerage fees. So that’s something less than 100%, oddly scaling down the more I have unrealized capital gains.
I also have financial assets in retirement accounts. Using those for payments (pre-retirement) would require paying a penalty. So that would be an even lower percentage.
(Again, I notice I am confused. The taxes in capital gains and any early-withdrawal penalty would be assessed later. So if my spouse needed life-saving surgery based on immediate payment, I can use all that money and go to jail later for failing to pay taxes. Is that an issue if access or willingness?)
I also have a current statutory right to old-age payments from social security when I get old enough. I have no way of accessing that now, so that gets a 0%. (Unless willingness to run up debt now against that future payment stream is relevant.)
Then there are illiquid assets. If I own a home, I can get a mortgage on it. That takes a while. Likewise, if I own a business, I can sell it. Or, if I have valuable but unused human capital, I can get a job or a higher paying job. (Note: it seems like a credit card is a way of accelerating access to job-based income. And other income.) All of those take time.
Now I notice that my approach of thinking about percentages isn’t doing everything needed. I need to know how much of an asset can be converted into access to payment. But I also need to think about time. Getting a mortgage might get me 75% of my home’s value, but only after a couple months.
So you could build a model of how much access I could have from all of these sources. The amount would presumably grow over the short to medium term as I could do things like convert illiquid assets to access.
But what does that tell us? In reality, I’m not going to mortgage up my house, so that possibility has very limited effects on prices. Perhaps I spend a little more of my immediate access every month knowing that my house is paid for. Perhaps we need yet another variable l, which I’ve been confused about -- something reflecting propensity or willingness to use the more difficult forms of access. Price levels might affect the willingness to use the access. It certainly does at the margin while I’m spending out of my checking account. On a graph where price leave is on the Y axis and my personal access at a moment in time is on the X axis, it would look like a 45 degree line from the origin that has asymptotically increasing slope. If we take a longer view, the slope would increase less quickly as I convert less liquid assets.
And now I’m out of steam without having reached a good conclusion. By my intuition is that one could still think about a meaningful quantity that drives the price level.
"But the government sent out checks to everyone, and those checks increased perceived wealth. If perceived wealth goes up and output goes down, the difference is bound to show up sooner or later in prices."
Sure, but this leaves out the Fed who is supposed to be on top of things like this, and lost of other things, to keep resetting its policy instruments (acting on perceived wealth?) to keep inflation (approximately) at a real income maximizing rate.
I'm guessing this is in a future essay, but I really like John Cochrane's explanation.
If you start with the demand for money, it's useful to separate the reasons for wanting money into usefulness in transactions and other reasons, which in a world without a gold standard is essentially just paying taxes.
Thus, as the world becomes more and more frictionless (cashless and digital), debt really does determine price level. Throw in a bunch of equations that I get lost in very quickly and you have a theory.
I'm not sure about all of this, but one feature of the "cashless society" that needs a lot of thought and addressing is missing. And that's the fact that a government can simply cut anyone off from the payment system if it feels like it (as in Canada last year).
A government doesn't even have to resort to overt violence if they can push a button and take away all of a protester's money.
This non-economist is always puzzled by the terms "cash" and "currency." Their meanings are derived from context. Last year I paid cash for a new pickup truck. I wrote a check, because I had enough cash in my checking account to cover it. At the gas station, I use a credit card. But to get the cash price, I would have needed physical currency. But "currency" is also vague and context-dependent. I hear there are foreign currency exchanges in which all the transactions are made digitally. When I was in the Congo/Zaire in the 1980s, the local currency was all nearly worthless paper. "Hard currency" referred to paper copies of US dollars. So I'm not sure what exactly is meant by a "cashless" or "cash free" society. In my current life, I can think of very few occasions where physical money is required or optimal. Perhaps a quarter for the neighborhood kids' lemonade? Or a $20 bill for the school fundraiser, only because the kid begging for it doesn't have a card reader? I did have a fellow drive to my house from about 120 miles away to buy my old pickup, and he had $16,000 in $100 bills for payment. But that's the kind of transaction, oddly enough, that will be flagged by the feds, because he must have withdrawn it from a bank account. (I dutifully reported the transaction to the state Dept. of Licensing, and put a copy in my tax file, just in case.) If they were counterfeit, he could have gone to prison for doing essentially what the treasury/fed does every microsecond. I'm sure there are day laborers being paid every day in physical money. But going cashless and having electronic payment transactions "on record" rather than transferring stacks of special pieces of paper printed with numbers would probably be better for all concerned.
I doubt your starting assumption that a doubling of the money supply will cause a doubling of prices; to me, that sounds like the "neutrality theory of money" disputed by *both* Hayek and Keynes (and many others).
Some people (such as those that sell things to the government) will notice the new money before others, so some prices will rise before others.
Your conclusions may still be valid, but its not persuasive to base them on a contested premise.
"Imagine that we had never relied on cash in order to trade. Instead suppose that we had always used computerized ledgers as the payments system." This is a really fun thought experiment. Filling in some important details. I assume that 1) there is a common unit of measure so that all accounts on all ledgers are denominated in the same "unit". Lets call it a unit. 2) The initial number of units had to be positive and is fixed. I have a strong intuition that the value of assets (measured in units) would be determined by the quanitity of units and according to your story, the price of things would as well.
Hmmm? 🤔
Milton Friedman on inflation:
old hat or “hat trick”?
"The main point that I hope you will consider is the possibility that in a cash-optional society access to the payment system is not controlled by raising or lowering the quantity of any particular asset that you can define as money."
I think this is going to take more explanation as I'm really struggling to see what's new here. I grew up in a small enough town that businesses had 'counter checks' from local banks that people would fill out so they didn't have to carry cash *or* physical checks, and I remember them disappearing as people got credit and ATM cards. Does anybody outside very small towns actually do old school charge accounts anymore? I can understand your illustration of wanting to have traveler's checks (I used to do that even when traveling in the US back in the day) versus leaving the country with nothing but a credit/ATM card (I've done the same thing several times in recent years). I just don't see this as something unique as much as it represents a simply a change in the form of access to the payment system, and especially that the payment system simply casts a wider net than it used to. An ATM swipe now transfers the money in your bank account the same way signing over a traveler's check did back in the day. There just wasn't an internet to perform the transfer back in 1970. I've been working since the 1980s and I've never received a paycheck other than by direct deposit.
I just don't think there's a good case to be made that the 'money supply' has ever been restricted to simply what the government calls currency in a fractional reserve banking system. When was the last time a bank gave a business physical cash as a loan? I don't even think that would have been common since the 1800s or even earlier.
I guess my point is that access to the payment system has never been restricted by currency supply except in the most extreme circumstances. I'm betting people who lived in company towns didn't have to carry much cash either but that wasn't a result of Fed policy.
Having liquid assets in tumultuous times is important and it is wise pointing out access to payment systems. This access is where real censorship and/or control can occur. This is why I believe there is value in having an allocation to Bitcoin. It is volatile, but it’s fixed supply should give it advantages over the long run in highly inflationary periods. It is also importantly a payment system. With the lightning network attached to Bitcoin, you can do instant peer to peer payments for a fraction of a penny across any distance. It will be interesting to see how the dynamics Arnold describes here interact with independent payment rails competing with traditional payment systems. Capital controls will be very tough in this new paradigm. Good luck everyone.
Wait a minute. Our current cash-optional system has cash and electronic equivalents, and they inter-convert through ATMs and banks. Most of the time it does not matter which you use for a transaction, unless you are doing something (like visiting a massage parlor) where you don't want a record kept of the transaction, or if you are using a business that refuses to take cash.
The price of cars, electronics, washers (& still allowed dryers rather than only sun), clothes & toys; plus both food and oil/energy, will continue to be priced more by competition & fighting over market share. Doubling income doesn't double the amount quantity of shoes bought (maybe Imelda?) nor the amount of food bought, tho it might shift quality preferences. But calling a $200 pair of shoes "twice as much shoes" as a $100 pair of shoes does not accurately capture the differences.
The non-asset inflation of increasing prices will level off, at a much higher price level, but then be more stable while we see, again, financial & house asset inflation. So that the rich get richer faster than the workers.
This occurs with or without cash.
Very interesting ideas about the differences going cash-less, but the much MUCH greater ability of gov't to control folk should make all freedom lovers oppose it.
Anarchotyranny. Almost here.
More questions
"When the pandemic hit, output of goods and services went down, because people were staying home. But the government sent out checks to everyone, and those checks increased perceived wealth. If perceived wealth goes up and output goes down, the difference is bound to show up sooner or later in prices."
I get that the pandemic hit on output (and related supply chain disruptions) was temporary even if in hindsight it was a little less so than many thought but I have a hard time understanding why Powell, Yellen, Krugman, and other supposedly knowing economists thought the increased spending, and resulting inflation, was short-term, transitory or otherwise not a continuing factor.
.
"If wealth were to suddenly jump by 10 percent, then after some period of time (say, ten years), spending will have been 10 percent higher than it would have been otherwise. Assuming that the actual path of goods and services produced is the same as what it would have been without the increase in perceived wealth, 10 percent more spending will translate into 10 percent higher prices than would have been experienced otherwise."
It took me a minute figure this one out. At least I think I did. It's an odd hypothetical. Wealth jumps 10% without a corresponding increase in production. It can happen but it seems more likely to confuse than clarify the piece. Do I have this right? … If prices increase 10%, was there really a 10% increase in wealth? … Which leads me to believe secondary effects note mentioned might be important here.
.
"As you know, I think that going forward the government needs to use inflation as the only means to deal with its mountain of debt. I expect that the Fed will have no power to stop this inflation."
I don't understand how this "deals with" debt other than maybe short term. Inflation causes interest on the debt to increase as much or more (eventually more). I'm not aware of any country with huge debt that came out of it without either a sacrifice to pay it down or a default.
.
"If the government doubles the money supply,"
This kind of misses an important part of what happens. It seems your whole piece does. While govt can add or remove money from the economy in a direct way, it also does it indirectly by changing interest rates it controls. That influences how much money banks and others will lend. Lending leverages the govt created money supply many times over.
.
"So that’s the best I can do to explain dollar prices."
I don't understand how "your best" says nothing about the velocity of money. That seems rather important to me.
Cash optional
"If we had always lived in a cash-optional or cashless society, then classical monetary theory would never have occurred to any economist."
As you say elsewhere, we currently live in a cash optional society. I don't get why this is different from it always having been this way. I don't get why it means we wouldn't operate the same under CMT if we'd always been cash optional, much less economists never would have thought of CMT. What am I missing?
Similarly, I don't get why using ledgers instead of cash makes any difference unless we also somehow remove money from the ledger system. My credit card and bank accounts are ledger systems that offer greater convenience for most but not all transactions but in the end it seems to me they do the exact same primary function as cash.
"In cash-optional society, access to the payment system is not determined by the quantity of money."
This seems true but I'm missing the point. It seems true for a system that is entirely cash based and most that are entirely cashless too. Is that incorrect?
"The main point that I hope you will consider is the possibility that in a cash-optional society access to the payment system is not controlled by raising or lowering the quantity of any particular asset that you can define as money."
Or maybe I have totally missed what you mean by cash optional. Do you mean dollar (currency) optional?
It sounds to me that you reflect or subscribe some of the views of John Cochrane and Scott Sumner?
It seems like access to the payment system is not a binary value. Instead, my access to the payment system is denominated in currency (even if I never hold currency).
I have a checking account with an ATM card attached to it. I can spend all of that, which counts as some number of dollars of access to the payment system. Presumably, the ratio of dollars in that account and access to the payment system is something like 100%. (I’m having trouble disentangling access from willingness to spend. For example, is the point where my balance falls low enough that I have to pay a monthly account fee relevant?)
I have a credit card with a limit. The credit limit seems to represent dollars of access to the payment system. Again, that seems like 100% access for the amount under my limit. In fact, I can probably exceed my limit, but that’s not 100% possible. Maybe there’s a formula there, like my bank would let me go 5% over before just shutting me off.
Then we get to sources of access that are clearly less than 100%.
I have financial assets that I can quickly convert to use for payments. Assume exchange traded funds outside a retirement account. I’d end up having to pay taxes in any capital gains, and there would be brokerage fees. So that’s something less than 100%, oddly scaling down the more I have unrealized capital gains.
I also have financial assets in retirement accounts. Using those for payments (pre-retirement) would require paying a penalty. So that would be an even lower percentage.
(Again, I notice I am confused. The taxes in capital gains and any early-withdrawal penalty would be assessed later. So if my spouse needed life-saving surgery based on immediate payment, I can use all that money and go to jail later for failing to pay taxes. Is that an issue if access or willingness?)
I also have a current statutory right to old-age payments from social security when I get old enough. I have no way of accessing that now, so that gets a 0%. (Unless willingness to run up debt now against that future payment stream is relevant.)
Then there are illiquid assets. If I own a home, I can get a mortgage on it. That takes a while. Likewise, if I own a business, I can sell it. Or, if I have valuable but unused human capital, I can get a job or a higher paying job. (Note: it seems like a credit card is a way of accelerating access to job-based income. And other income.) All of those take time.
Now I notice that my approach of thinking about percentages isn’t doing everything needed. I need to know how much of an asset can be converted into access to payment. But I also need to think about time. Getting a mortgage might get me 75% of my home’s value, but only after a couple months.
So you could build a model of how much access I could have from all of these sources. The amount would presumably grow over the short to medium term as I could do things like convert illiquid assets to access.
But what does that tell us? In reality, I’m not going to mortgage up my house, so that possibility has very limited effects on prices. Perhaps I spend a little more of my immediate access every month knowing that my house is paid for. Perhaps we need yet another variable l, which I’ve been confused about -- something reflecting propensity or willingness to use the more difficult forms of access. Price levels might affect the willingness to use the access. It certainly does at the margin while I’m spending out of my checking account. On a graph where price leave is on the Y axis and my personal access at a moment in time is on the X axis, it would look like a 45 degree line from the origin that has asymptotically increasing slope. If we take a longer view, the slope would increase less quickly as I convert less liquid assets.
And now I’m out of steam without having reached a good conclusion. By my intuition is that one could still think about a meaningful quantity that drives the price level.
"But the government sent out checks to everyone, and those checks increased perceived wealth. If perceived wealth goes up and output goes down, the difference is bound to show up sooner or later in prices."
Sure, but this leaves out the Fed who is supposed to be on top of things like this, and lost of other things, to keep resetting its policy instruments (acting on perceived wealth?) to keep inflation (approximately) at a real income maximizing rate.
Sounds a lot like you are in concert, possibly melding the works of John Cochrane and Scott Sumner.
I'm guessing this is in a future essay, but I really like John Cochrane's explanation.
If you start with the demand for money, it's useful to separate the reasons for wanting money into usefulness in transactions and other reasons, which in a world without a gold standard is essentially just paying taxes.
Thus, as the world becomes more and more frictionless (cashless and digital), debt really does determine price level. Throw in a bunch of equations that I get lost in very quickly and you have a theory.