If you have 100 bitcoins in your computer wallet and 100 in your MtGox account,

[BurtW]

Atheros,

you write on the wiki page:

There is disagreement concerning whether Fractional Reserve Banking is realistically possible with Bitcoin, and what it would require.

I have said multiple times that I do not claim that FRB with Bitcoin is impossible. On the contrary, I provided several historical examples of bitcoin FRB. Why do you keep misrepresenting my claim?

I did not put enough thought into my previous response to this post and it ended up totally wrong.  So I have deleted it and here is another attempt at a summary:

There is general agreement that FRB is theoretically possible.  Anyone could set up a bank, convince people to deposit their hard earned BTC into it (pay interest, give away toasters, etc.) and then loan out a percentage of these BTC that they recieve - keeping a fraction on reserve to handle day to day BTC requests from their customers that have BTC deposited at the bank.

However there is strong disagreement as to how the existence of BTC FRB might affect the BTC money supply.

[lonelyminer (Peter Šurda)]

Note that RB are not fungible. They're always issued from one person to another.

For simplicity, I was assuming that there is only one RB issuer. However, I agree that this is not a necessary requirement. Just like in Northern Ireland, Scotland or Hong Kong there are multiple bank note issuers, there could be also multiple RB issuers. To what extent they would be treated as fungible is of course another question.

Anyway, you're considering ripple a money substitute in the austrian sense, so maybe we should take it out of our example.

I do not have a theoretical objection to any debt instrument becoming a money substitute. The question is the likelihood. It needs to offer something as a method of payment which the money-proper does not. With gold/fiat, this is straightforward. With Bitcoin, not so much. Even the hypothesised advantages of ripple might not be sufficient to cause enough demand, I just think it makes more sense than with classical demand deposits.

Let's just use regular credit transactions. I meant money substitutes as just other medium of exchange different from money, don't know if this fits with the austrian definition.

In Austrian terminology, as far as I know, money-substitute is a debt claim on money-proper which is generally accepted as a medium of exchange instead of money.

"We said that wares represent a demand for the medium of exchange exactly corresponding to their amount and quality. So, if there were any method of exchanging wares without employing money, the demand for money would be reduced by the amount of the wares so exchanged. This is self-evident when examined with the aid of our conception of the demand for money. Here again we may use a railway-line as an illustration. The demand for rolling stock is exactly equal to the amount of goods awaiting transport. But if a canal is built along the railway, the demand for rolling stock decreases by the amount of the goods transported by canal."

I've heard about Gesell, I read Alternatives to Legal Tender (or something like that) by Thomas Greco. I agree with the quote, however it neglects the informational function of money, i.e. the intertemporal and interpersonal coordination of plans, as well as the speculation (uncertain future). These needs might affect the outcome.

When I said inflation I meant price inflation rather than monetary inflation (what you defined as inflation). I don't care if credit it's going to be considered part of M or not. What I want to know is if prices would rise.

This is a slightly controversial topic in the Austrian circles. The main problem, as far as I understand it, is not really whether prices would rise (they might not, assuming the extra supply is offset by extra growth of the economy), but that the new money is "injected" into the economy without causing corresponding production costs. This causes a redistribution of purchasing power as the new money propagates through the economy, and leads to boom & bust cycles (Austrian Business Cycle Theory).

I'm interested in both the Austrian and the Keynesian view on this.

I am not an expert in the Keynesian school, but I think Keynes argued that if there is involuntary unemployment, expansion of money supply does not cause inflation.

If we had only one country with one currency (say gold) and 100% reserve for banks by law, will private credit between partners push price inflation?

Under typical circumstances, no. This is actually how the full reservist branch of the Austrian school sees a proper economy.

[lonelyminer (Peter Šurda)]

I did not put enough thought into my previous response to this post and it ended up totally wrong.  So I have deleted it and here is another attempt at a summary:

Ok.

There is general agreement that FRB is theoretically possible.  Anyone could set up a bank, convince people to deposit their hard earned BTC into it (pay interest, give away toasters, etc.) and then loan out a percentage of these BTC that they recieve - keeping a fraction on reserve to handle day to day BTC requests from their customers that have BTC deposited at the bank.

I agree, merely would like to add that

• by this debt-instrument circulating, the risk of performing FRB is decreased and becomes more likely
• term deposits or other investment instruments, which do not have zero maturity, should not be considered relevant for this topic

I hope none of my dissenters will object to these two points.

However there is strong disagreement as to how the existence of BTC FRB might affect the BTC money supply.

I also agree.

[jtimon]

I think you're not getting my point. I think I don't mean money substitutes the way you do.
I'm talking about credit from the provider to the purchaser of the good.

I'll put here another quote also from the same text.

As credit transactions have in this way a powerful influence upon the demand for money, we must consider them somewhat more closely.

 We said that wares represent a demand for the medium of exchange exactly corresponding to their amount and quality. So, if there were any method of exchanging wares without employing money, the demand for money would be reduced by the amount of the wares so exchanged. This is self-evident when examined with the aid of our conception of the demand for money. Here again we may use a railway-line as an illustration. The demand for rolling stock is exactly equal to the amount of goods awaiting transport. But if a canal is built along the railway, the demand for rolling stock decreases by the amount of the goods transported by canal.

If this is true, an increase in private credit or just an increase in bartering would cause price inflation (assuming zero growth to simplify).
Those things could also happen under a gold standard. So, even if cycles don't get worse through elastic monetary policies, the inflation/deflation cycles may also appear within an unelastic monetary supply. Of course, this assumes a money monopoly, which will never be the case for bitcoin.

You say that the money substitute should have an advantage over money to be used. I disagree. Sometimes the lack of money is enough of an advantage. What's the advantage for B when paying with IOUs instead of money? Well, he doesn't have the money. What is the advantage for C in accepting it? None, only a higher price.

In summary, the sentence by Gesell I would like you to agree with or to deny is:

"The daily demand for money therefore equals the quantity of wares daily brought to market, less the wares exchanged by way of credit (or barter)."

Note that when he says wares he includes all produced goods and services.

[lonelyminer (Peter Šurda)]

If this is true, an increase in private credit or just an increase in bartering would cause price inflation (assuming zero growth to simplify).

I think I get what you mean. You're talking about the substitution effect of credit. But unless the this private credit circulates, this is actually a shrinking of the economy, because the creditor is withdrawing his activites from the bigger market. If the creditor had a choice, he would not act this way because it's detrimental for him. I think this is the problem with the proposals of voluntary local debt-based systems like those Gesell proposes.

Those things could also happen under a gold standard. So, even if cycles don't get worse through elastic monetary policies, the inflation/deflation cycles may also appear within an unelastic monetary supply. Of course, this assumes a money monopoly, which will never be the case for bitcoin.

I agree. Inelastic money supply does not prevent all errors, just the systematic ones. Most of the issues only arise due to monopolies.

You say that the money substitute should have an advantage over money to be used. I disagree. Sometimes the lack of money is enough of an advantage. What's the advantage for B when paying with IOUs instead of money? Well, he doesn't have the money. What is the advantage for C in accepting it? None, only a higher price.

What you describe means that there is some sort of interference with the market. If a seller has the choice between selling for money and selling for credit, why would he accept credit? Merely because someone has an IOU does not mean anyone would accept this IOU. At par even.

In summary, the sentence by Gesell I would like you to agree with or to deny is:

"The daily demand for money therefore equals the quantity of wares daily brought to market, less the wares exchanged by way of credit (or barter)."

Note that when he says wares he includes all produced goods and services.

I think this ignores investing (people selling future goods), but for simplicity we can leave it as it is. Otherwise I don't really have a problem with this formulated this way. I only worry that this leads to erroneous conclusions, much like the Keynesian C + I + G = Y leads to the assumption that by increasing G, one is increasing Y.

[Atheros]

Atheros, I find it odd how you are presenting a simplification that ignores my point as as a refutation.

Your point that money supply would not include demand deposits if the demand deposits don't include checking services or EFTs? How could I prove a negative? How could I find a quote which directly says that that isn't the case? I can't. If someone were to assert that demand deposits are only included in the money supply if the depositor's last name starts with a vowel, I won't be able to find a quote to counter that claim. I'll only be able to explain why that definition isn't useful, as I have tried to do.

(I interpret this as liquidity being the determining factor for money)

I agree.

The complex of rates of interest would simply be an expression of the terms on which the banking system is prepared to acquire or part with debts; and the quantity of money would be the amount which can find a home in the possession of individuals who — after taking account of all relevant circumstances — prefer the control of liquid cash to parting with it in exchange for a debt on the terms indicated by the market rate of interest.

(I interpret this as Keynes not considering deposits a part of the "quantity of money", i.e. what we call money supply).

I disagree because demand deposits are highly liquid just like cash. They also do not earn "the market rate of interest" that a normal "debt" would earn.

Concerning Keynes, I might not be getting FRB right either. "The General Theory of Employment, Interest
and Money" doesn't reference Fractional Reserve anything or Money Supply. I would have called my viewpoint the Mainstream Viewpoint, but I figured you would have objected.

As a side remark, I find it funny how, allegedly, both Austrians and Keynesians see Bitcoin as doing exactly what they want, even though these goals are opposed.

I think that Keynesian economists like Krugman would and do dislike Bitcoin (as well as a Gold standard) from an economic-good perspective and I think that they are justified. High deflation is terrible for an economy. Fortunately we aren't discussing the theoretical good or harm to the economy- we are mostly discussing fractional reserve banking. 

[Atheros]

There is general agreement that FRB is theoretically possible.

I contend that FRB is not just theoretically possible, it is possible and practical.

I agree, merely would like to add that

• by this debt-instrument circulating, the risk of performing FRB is decreased and becomes more likely
• term deposits or other investment instruments, which do not have zero maturity, should not be considered relevant for this topic

I can't tell what the first one means. What is this debt-instrument? You talk about debt instruments in your section but I do not. Can we thus not mention it in the introduction?

Concerning the second bullet, lets just leave it out if neither of us are talking about it.

"However there is strong disagreement as to how the existence of BTC FRB might affect the BTC money supply."

Did anyone disagree on this? Fractional reserve banking necessarily increases the money supply. Did someone disagree?

We all agreed that FRB with Bitcoin is possible, we just disagreed on what it would entail, right? So why not just say that?

While Fractional Reserve Banking with Bitcoin is possible, there is much disagreement about what it would entail.

[Atheros]

When I said inflation I meant price inflation rather than monetary inflation (what you defined as inflation). I don't care if credit it's going to be considered part of M or not. What I want to know is if prices would rise.
I'm interested in both the Austrian and the Keynesian view on this.

If we had only one country with one currency (say gold) and 100% reserve for banks by law, will private credit between partners push price inflation?

My View, which is probably Keynesian:
Fractional Reserve Banking increases the money supply, which causes inflation. Inflation is the rise in the general prices of goods and services in an economy over a period of time. The inflation would be limited because the money supply could only rise to a certain multiple of the monetary base. For example, the money supply could rise to two times the monetary base, but wouldn't go any higher.

Private credit wouldn't push inflation because it is not the case that two people both believe they have easy access to the same bitcoins; one person has the bitcoins and the other doesn't and they know it. IOUs are a little different; instead of lending you bitcoins, the lender lends you a product or service and expects payment back in bitcoins. OR you could say that he lends you bitcoins, and then pays himself with those bitcoins. Although the IOU is denominated in BTC, neither of you are actually using bitcoins yet. Thus the demand for BTC will be lower until you sell something to acquire BTC to repay the IOU. The IOU would lower the value of bitcoins by a very small amount: the same as if the two parties had decided to not use BTC in the first place.

If millions of people today started writing IOUs denominated in bitcoins which all required that they be repaid in a month, the value of the dollar would drop a little bit because people would have less demand for dollars. The value of individual bitcoins would stay the same for much of the month until people started acquiring bitcoins to repay the IOUs, at which point the value of each bitcoin would skyrocket.
(I'm ignoring the effects of instantaneously informing millions about Bitcoin for the sake of the example.)

[lonelyminer (Peter Šurda)]

Your point that money supply would not include demand deposits if the demand deposits don't include checking services or EFTs? How could I prove a negative? How could I find a quote which directly says that that isn't the case? I can't. If someone were to assert that demand deposits are only included in the money supply if the depositor's last name starts with a vowel, I won't be able to find a quote to counter that claim.

So, according to your logic, because you cannot find a book that says that plasma does not exist, a high school text book claiming there are only three states of matter refutes specialist publications that mention plasma. This is what your argument boils down to.

I'll only be able to explain why that definition isn't useful, as I have tried to do.

I explained that your explanation does not match any economic theory that I know of and invalidates the concepts of money velocity and/or GDP. In your example, even though people cannot buy things that are more expensive than the money base but less than the value of the debt instruments, you nevertheless claim that the money supply corresponds to the higher value. What's the use of your definition? There is none. It cannot be integrated into any economic theory.

I disagree because demand deposits are highly liquid just like cash. They also do not earn "the market rate of interest" that a normal "debt" would earn.

I worry you conflate liquidity and maturity. If you can only redeem the deposit at the issuer, why does it affect the supply? That's like saying that if I issue a zero-maturity debt instruments denominated in tomatoes, regardless of whether this instrument is accepted by the general public instead of actual tomatoes, it increases the supply of tomatoes. This makes no sense. You invented a new economic theory apparently.

Concerning Keynes, I might not be getting FRB right either. "The General Theory of Employment, Interest
and Money" doesn't reference Fractional Reserve anything or Money Supply. I would have called my viewpoint the Mainstream Viewpoint, but I figured you would have objected.

The term "money supply" was popularised by the monetarists in the 60s. Keynes used "quantity of money" or "money quantity".

I think that Keynesian economists like Krugman would and do dislike Bitcoin (as well as a Gold standard) from an economic-good perspective and I think that they are justified. High deflation is terrible for an economy. Fortunately we aren't discussing the theoretical good or harm to the economy- we are mostly discussing fractional reserve banking. 

I agree that Keynesians probably dislike Bitcoin. I disagree that deflation is bad. See e.g. Deflation and Liberty by Joerg Guido Huelsmann

[lonelyminer (Peter Šurda)]

I contend that FRB is not just theoretically possible, it is possible and practical.

Your description, however, does not accurately reflect FRB as we know it. What you portray is more similar to an investment in general.

I can't tell what the first one means. What is this debt-instrument? You talk about debt instruments in your section but I do not. Can we thus not mention it in the introduction?

Debt instrument is a generic name for an obligation denominated in the unit. Demand deposits, cheques, EFT, bank notes based on gold, personal credit, these are all debt instruments. I introduced it to make my argument more accurate.

Concerning the second bullet, lets just leave it out if neither of us are talking about it.

Excellent.

Did anyone disagree on this? Fractional reserve banking necessarily increases the money supply. Did someone disagree?

As I explained, money supply is the amount of money base plus debt instruments in circulation. But this is not a superset of FRB. These are two sets which contain a certain overlap.

[lonelyminer (Peter Šurda)]

My View, which is probably Keynesian:
Fractional Reserve Banking increases the money supply, which causes inflation.

Keynes claims that as long as there is involuntary unemployment, an increase in the quantity of money (which is what he called money supply) does not cause inflation.

Private credit wouldn't push inflation because it is not the case that two people both believe they have easy access to the same bitcoins; one person has the bitcoins and the other doesn't and they know it. IOUs are a little different; instead of lending you bitcoins, the lender lends you a product or service and expects payment back in bitcoins. OR you could say that he lends you bitcoins, and then pays himself with those bitcoins. Although the IOU is denominated in BTC, neither of you are actually using bitcoins yet. Thus the demand for BTC will be lower until you sell something to acquire BTC to repay the IOU. The IOU would lower the value of bitcoins by a very small amount: the same as if the two parties had decided to not use BTC in the first place.

The IOU has no effect on the money supply. The effect you describe is not an increase of the money supply, it is a decrease in the size of the Bitcoin economy, because there is a good that is not traded against Bitcoins anymore. The value of Bitcoins would only decrease if the change in size was not compensated by a change in the money velocity (in your example, it might or might not, we don't know).

If millions of people today started writing IOUs denominated in bitcoins which all required that they be repaid in a month, the value of the dollar would drop a little bit because people would have less demand for dollars. The value of individual bitcoins would stay the same for much of the month until people started acquiring bitcoins to repay the IOUs, at which point the value of each bitcoin would skyrocket.
(I'm ignoring the effects of instantaneously informing millions about Bitcoin for the sake of the example.)

The effect you describe is an outflow of goods from the "dollar economy" into the "Bitcoin IOU economy", and subsequently into the "Bitcoin economy". At the first stage, the size of the Bitcoin economy would remain constant, because there's no change in the supply or demand of Bitcoins. At the end of the month, the demand for Bitcoins would increase while the supply would be the same, thus the price would increase.

I'm still not sure though what this has to do with your position.

[rahl]

But if I can demand the depot back easily at any time, how is it functionally any different? Why would I not view myself as owning 200 bitcoins seeing as I can access 200 bitcoins easily?

As long as the trust and reliability of if the bank is high enough yes.

First of all it is not fractional reserve any longer or a demand deposit if you have any kind of contractual restriction on redemption (well at least that isn't purely logistical as complying with the opening hours of the money warehouse). If you got some kind of restriction that let's the bank lend out the money you don't have a demand deposit you have a IOU, even if the bank actually repay it's debt instantly in 100% of cases.

Now if everyone believes the bank will be able to pay instantly in 100% of cases then it is likely those particular IOU will become as good as money and the money supply will increase anyway. But this should all still be ok since the price of money will still be correct and the market can account for trust and reliability of various banks.

But should actually not really happen, there is always extra risk of an IOU compared to money. So they should be traded at a discount. Just that the transactional fees of counting the discount might be higher then the discount if the bank is incredibly effective. Anyway the important thing is that the market can distinguish between money and IOUs and account for it accordingly which it can not do in a central banking system, in a system where banks lie or in a system where warehouse banking have special legal advantages over other warehouse storage services.

This could be a problem with Bitcoin at this stage because the institutions to prevent the lying like auditing is not really there. But then again who would an exchange be lending there Bitcoin too that wouldn't make there operation incredibly risky for them anyway?
I think it is fairly safe to assume all the Bitcoin are in there wallet. If they are running fractional practices in anything it is more likely to be USD...

[lonelyminer (Peter Šurda)]

rahl, since you have Mises as an avatar, can you review http://en.bitcoin.it/wiki/Fractional_Reserve_Banking_and_Bitcoin and comment on it?

This could be a problem with Bitcoin at this stage because the institutions to prevent the lying like auditing is not really there. But then again who would an exchange be lending there Bitcoin too that wouldn't make there operation incredibly risky for them anyway?
I think it is fairly safe to assume all the Bitcoin are in there wallet. If they are running fractional practices in anything it is more likely to be USD...

During my research of Bitcoin, I asked genjix from Intersango and Darren from Tradehill about FRB. They said that they're not conducing FRB. Genjix was particularly opposed to FRB and said something like "it's not my money".

[Atheros]

Your point that money supply would not include demand deposits if the demand deposits don't include checking services or EFTs? How could I prove a negative? How could I find a quote which directly says that that isn't the case? I can't. If someone were to assert that demand deposits are only included in the money supply if the depositor's last name starts with a vowel, I won't be able to find a quote to counter that claim.

So, according to your logic, because you cannot find a book that says that plasma does not exist, a high school text book claiming there are only three states of matter refutes specialist publications that mention plasma. This is what your argument boils down to.

Thinking Mises is the 'specialist publication' and the Fed's website is the high school textbook is lunacy.

I'll only be able to explain why that definition isn't useful, as I have tried to do.

I explained that your explanation does not match any economic theory that I know of and invalidates the concepts of money velocity and/or GDP. In your example, even though people cannot buy things that are more expensive than the money base but less than the value of the debt instruments, you nevertheless claim that the money supply corresponds to the higher value. What's the use of your definition? There is none. It cannot be integrated into any economic theory.

I won't be able to respond to what you say unless you speak more precisely. Nevermind. Don't care.

[lonelyminer (Peter Šurda)]

Thinking Mises is the 'specialist publication' and the Fed's website is the high school textbook is lunacy.

I also quoted wikipedia and some other guy I found on the internet. Ignorance is knowledge.

I won't be able to respond to what you say unless you speak more precisely. Nevermind. Don't care.

Ignorance is knowledge.

[lonelyminer (Peter Šurda)]

Macroeconomics: Private and Public Choice By James D. Gwartney, Richard L. Stroup, Russell S. Sobel, David MacPherson (2010):

The M1 Money Supply
Above all else, money is a medium of exchange. The narrowest definition of the money supply, M1, focuses on this function. Based on its role as a medium of exchange, it is clear that currency - coins and paper bills - falls into this definition. But currency isn't the only form of money readily used for exchange. If you want to buy something from a store, many will let you pay with either a check or debit card that will transfer funds from your bank account to theirs. Therefore, checkable bank deposits that can easily be used as a means of payment should be included in the M1 money supply measure.

(emphasis added)
Not available online, but google books can provide this quote.

Ralph Byrnes (University of North Carolina): An illustrated encyclopedia of economics, 2009:
http://www.unc.edu/depts/econ/byrns_web/Economicae/moneysupply.html

Currency (coins and paper bills) is the most easily identified component of the money supply because it may be used (a) for virtually all transactions, (b) to price goods and services, and (c) as an asset. Demand deposits are the other major assets that perform all monetary functions.
...
Demand deposits and currency are the only major assets that are both mediums of exchange and widely accepted as money.

(emphasis added)

Encyclopedia of Business http://www.referenceforbusiness.com/encyclopedia/Mor-Off/Money-Supply.html

One can now see how all three components of the M1 measure of money supply—currency, checkable deposits and traveler's checks—are clearly money in the ordinary sense of the term, as they can be directly used in payment of goods and services.

(emphasis added)

But please Atheros, feel free to deny and ignore.

[johnyj]

I think FRB based on a simple fact that people do not utilize their saving in the bank until much later.

It works for 2-3 generations when savings are expanding, but when a country has matured and shifted to older populations, more people will be taking money out of the banks, FRB will have big trouble. In principle, all the saved money will be spend sooner or later, banks can not keep the same lending speed unless the growth/inflation is enough high to offset the slow down in saving

[Atheros]

In response to lonelyminer,
Those quotes are good and clear. So if Bitcoin demand deposits are not in M1, they would certainly be in M2. And Fractional Reserve Banking can still be easily exercised with funds counted in M2. No substitutes-circulating-as-money are required.

You now have my admission that Bitcoin demand deposits might not be counted as M1, but rather M2.

In response to johny,
What do people do with money they take out of a savings account. Spend it, right? Then what do the people who receive the money do? Either spend it further or put it back in the bank.

[johnyj]

In response to johny,
What do people do with money they take out of a savings account. Spend it, right? Then what do the people who receive the money do? Either spend it further or put it back in the bank.

In my opinion, those saved money is just an illusion, there is no corresponding wealth to consume when they start to spend those savings, so inflation is the first thing to come

But this is not very precise, I will try to use an island model to explain this, will be much easier

[lonelyminer (Peter Šurda)]

In response to lonelyminer,
You now have my admission that Bitcoin demand deposits might not be counted as M1, but rather M2.

Oh, progress, there she is, I found you at last.

I admit that it is a bit less clear with Bitcoin demand deposits in M2 than in M1, but I still don't think it qualifies. Please bear with me for a minute while I formulate my argument.

The last reference I posted says this about M2:

One can assert that the M2 measure of money supply is a somewhat lower on the liquidity scale, compared to the M1 measure, which is the most liquid. Assets that are added to the M1 aggregate to arrive at the M2 measure are items that are the most easily and most frequently transferred into checking accounts or can otherwise be converted into cash quickly.

Saving accounts or term deposits (typical examples of M2) can be converted into demand deposits at full nominal value. M2 stays the same, while M1 increases. If everyone did this at the same time, while it might cause disruptions on the markets if overissue was present, it is a possible course of action. With Bitcoin, this is not possible, because the amount is limited by the reserves.

The example of M2 at the wikipedia page on Money Supply:

Laura writes check number 7774 for $1000 and brings it to the bank to start a Money Market account (these do not have a credit-creating charter), M1 goes down by $1000, but M2 stays the same. This is because M2 includes the Money Market account in addition to all money counted in M1.

seems to match the pattern.

In other words, I think that M2 and M3 only exist because typically, demand deposits are in M1. Since this is not the case with Bitcoin, I therefore conclude that Bitcoin demand deposits would not be in M2 or M3 either. I think that unless you can inflate M1 over the amount of reserves, M1 = M2 = M3. While people might think that M2/M3 functions as a store of value, if it neither acts as a medium of exchange, nor, nor can it be converted into one in an amount exceeding the reserves, it should not have an effect on the supply.

However, I admit there is a greater level of difficulty here and a more thorough analysis is necessarly.