The Bitcoin deflation annoyance

[Blinken]

Despite the current trend of Bitcoins losing value against the dollar I think long term the greater annoyance is going to be a constant rise in value. As more crypto currency is used its value will increase and since the quantity of coins is limited at 21 million the value of each coin will keep increasing. This will be annoying for two reasons:

(1) the prices of things will be constantly changing since the currency will keep deflating

(2) we will be dealing in decimal units. Quoting ridiculous prices like BC0.000013 or whatever will be very annoying and a lot of invoicing systems are not designed to handle cents values below 1/100

I think ultimately we will have to make a competing client that breaks the 21 million limit and uses a more practical value scale and rate of inflation.

[1714135220]

1714135220

[RodeoX]

This has been talked about extensively, and you are not alone in your opinion. However I would personally have no interest in a currency with such controls. I prefer to think of BTC like gold. With a fixed supply and a variable demand.

[evoorhees]

(1) the prices of things will be constantly changing since the currency will keep deflating

That's true of the US Dollar as well, but instead of the money losing purchasing power it will gain purchasing power. The rate of change will be driven lower by speculators long term (if everyone knows Bitcoins will rise in price, speculators will buy them, rising the price in the present). Over time, you will see a gradual, relatively predictable price and its annoyance for repricing goods should not be overwhelming. How many times per day does Wal Mart change the price on a good in their store? It's no great burden.

(2) we will be dealing in decimal units. Quoting ridiculous prices like BC0.000013 or whatever will be very annoying and a lot of invoicing systems are not designed to handle cents values below 1/100

Have some foresight!  A name will be given to smaller units. Perhaps the .00000001 unit will be called the Satoshi (seems to be catching on). In such a case where BTC is extremely valuable, your coffee will simply be "5.30 satoshis".   Nomenclature is a non-issue, the market will easily figure out efficient naming schemes.

I think ultimately we will have to make a competing client that breaks the 21 million limit and uses a more practical value scale and rate of inflation.

Go for it. I won't be buying it.  

[tvbcof]

Despite the current trend of Bitcoins losing value against the dollar I think long term the greater annoyance is going to be a constant rise in value. As more crypto currency is used its value will increase and since the quantity of coins is limited at 21 million the value of each coin will keep increasing. This will be annoying for two reasons:

(1) the prices of things will be constantly changing since the currency will keep deflating

price volatility is hardly an unheard of phenomenon in any currency system.

(2) we will be dealing in decimal units. Quoting ridiculous prices like BC0.000013 or whatever will be very annoying and a lot of invoicing systems are not designed to handle cents values below 1/100

They can adapt or die.  It isn't rocket science.  Indeed, having decimal places already makes the jump that much easier.  It's not like (most) people are forced into using base-e or something.

I think ultimately we will have to make a competing client that breaks the 21 million limit and uses a more practical value scale and rate of inflation.

The more the merrier in my opinion.  People can choose which one suites them and exchanges can do a good business automatically adjusting for people so they won't need to care that much, though they could save the exchange fee by operating within their chosen block-chain.  My hope is mostly that each alternate chain is backed by value in the 'satoshi block-chain'.

[Blinken]

This has been talked about extensively, and you are not alone in your opinion. However I would personally have no interest in a currency with such controls. I prefer to think of BTC like gold. With a fixed supply and a variable demand.

Gold does not have a fixed supply. Over 2,000 tonnes are mined every year, increasing the supply by at least 1% annually if not more.

[Steve]

(1) the prices of things will be constantly changing since the currency will keep deflating

Just set your prices relative to something else you find desirable (or perhaps your input costs) and let software do the updating for you.

[Etlase2]

(1) the prices of things will be constantly changing since the currency will keep deflating

...

I think ultimately we will have to make a competing client that breaks the 21 million limit and uses a more practical value scale and rate of inflation.

While far in the future deflation would happen on a slow timescale, I think you might prefer a system where the currency strives to maintain the same buying power. See my signature.

[BitterTea]

This has been talked about extensively, and you are not alone in your opinion. However I would personally have no interest in a currency with such controls. I prefer to think of BTC like gold. With a fixed supply and a variable demand.

Gold does not have a fixed supply. Over 2,000 tonnes are mined every year, increasing the supply by at least 1% annually if not more.

There is a finite amount of gold available for us to extract out of the earth, so yes, there is a fixed supply. We just know exactly when we'll run out of bitcoins to mine, whereas we can only guess at when we'll run out of economically available gold deposits. We've already mined all of the easy gold, it will only get more and more expensive to mine gold as time goes on.

[DeathAndTaxes]

Simple solutions.

1) Price of everything is always changing.  Hell prices don't stop changing.  That is a complete non-issue.  Currently BTC has extreme volatility but that has nothing to do with deflation.

2) If we ever reached a point where one BTC was worth say $82,000 likely people would simply price things in Satoshi's.  


Likely most people would price everyday items in satoshis.  Large items could either be priced in BTC or BTC can become completely arhaic and people would simply express things in thousands, millins, billions of satoshis.

For example (@ $82,000 USD per BTC):
$5 fastfood meal = 0.000060976BTC or 6100 Satoshis.
$80,000 salary = 0.9756 BTC or 97.5 million Satoshis.
$3 mil company = 36.58 BTC or 3.6 billion Satoshis.

Alternatively SI prefixes could be used:
mBTC (microBTC) = 1/1000th BTC =         0.001 BTC
μBTC (miliBTC) = 1/1,000,000 BTC =         0.000001 BTC
nBTC (nanoBTC) = 1/1,000,000,000 BTC = 0.0000000001 BTC

Terminalogy will adapt to pricing in alternate units.  Of all things to worry about related to Bitcoin this is a complete non-issue.

[Blinken]

(1) the prices of things will be constantly changing since the currency will keep deflating

...

I think ultimately we will have to make a competing client that breaks the 21 million limit and uses a more practical value scale and rate of inflation.

While far in the future deflation would happen on a slow timescale, I think you might prefer a system where the currency strives to maintain the same buying power. See my signature.

Dude, your thing requires registration to read. There is no way I going to get tracked just to read your thing.

[memvola]

There is an immense amount of discussion in this forum regarding your points, so you'd need to focus on specifics in order to avoid repeating what others have said. Also, money supply inflates, so Bitcoin is "constantly" deflationary only if you assume a steeper adoption pattern, which is an unknown.

(1) Prices are always changing. We are talking about a massive adoption scenario, so in usual circumstances, prices shouldn't change faster than fiat currencies or even gold. Also, inflation or deflation doesn't matter in this case, so it's more a matter of volatility than deflation/inflation.

(2) Bitcoin's value will not increase 1 million-fold in just a year, or even a decade, so there is enough time for you to decide on a different unit. Use uBTC.

There is nothing to be done practically about matters that are unknown. These issues are pretty minor if you think about the timescale they are relevant. Instead of trying to predict the future, adopting a pattern that is easier to conceptualize was a better choice.

Gold does not have a fixed supply. Over 2,000 tonnes are mined every year, increasing the supply by at least 1% annually if not more.

So is it linear? We don't know. Have we reached peak gold? Will we discover a new way to mine it? These are all unknowns. There is definitely a fixed amount of gold we can mine within a practical timescale though. What we know, is that gold is a preferred medium for storing value because it is scarce and this will not change. It would the perfect medium if supply wasn't increasing.

We don't know the future of Bitcoin either. It would be undesirable to have a generation function that would get in a feedback loop with the price, even if it were possible. So the only option is to derive a function from the desired traits of the currency.

[Etlase2]

Dude, your thing requires registration to read. There is no way I going to get tracked just to read your thing.

mediafire doesn't require registration, I didn't even have to register to upload. And googledocs certainly doesn't.

[JoelKatz]

Despite the current trend of Bitcoins losing value against the dollar I think long term the greater annoyance is going to be a constant rise in value. As more crypto currency is used its value will increase and since the quantity of coins is limited at 21 million the value of each coin will keep increasing. This will be annoying for two reasons:

The effective quantity of Bitcoins is not limited to 21 million. Loans increase the effective quantity of Bitcoins. Alternative blockchains increase the effective quantity of Bitcoins. It won't be a problem -- unless you imagine a powerful government that prohibits all business practices the increase the effective monetary supply, the effective monetary supply will grow as the economy does.

For example, imagine we're in a bitcoin-based economy. I want to buy a car that costs 5,000 bitcoins, but I don't have 5,000 bitcoins. I borrow 5,000 bitcoins and in exchange give an IOU worth about 5,000 bitcoins. The 5,000 bitcoins I borrowed are still in the economy, being spent by the guy I got the car from. But the IOU is also in the economy, and it effectively adds an additional 5,000 bitcoins because it is interchangeable in the marketplace with 5,000 bitcoins.

[Blinken]

There is an immense amount of discussion in this forum regarding your points, so you'd need to focus on specifics in order to avoid repeating what others have said. Also, money supply inflates, so Bitcoin is "constantly" deflationary only if you assume a steeper adoption pattern, which is an unknown.

(1) Prices are always changing. We are talking about a massive adoption scenario, so in usual circumstances, prices shouldn't change faster than fiat currencies or even gold. Also, inflation or deflation doesn't matter in this case, so it's more a matter of volatility than deflation/inflation.

(2) Bitcoin's value will not increase 1 million-fold in just a year, or even a decade, so there is enough time for you to decide on a different unit. Use uBTC.

There is nothing to be done practically about matters that are unknown. These issues are pretty minor if you think about the timescale they are relevant. Instead of trying to predict the future, adopting a pattern that is easier to conceptualize was a better choice.

Gold does not have a fixed supply. Over 2,000 tonnes are mined every year, increasing the supply by at least 1% annually if not more.

So is it linear? We don't know. Have we reached peak gold? Will we discover a new way to mine it? These are all unknowns. There is definitely a fixed amount of gold we can mine within a practical timescale though. What we know, is that gold is a preferred medium for storing value because it is scarce and this will not change. It would the perfect medium if supply wasn't increasing.

We don't know the future of Bitcoin either. It would be undesirable to have a generation function that would get in a feedback loop with the price, even if it were possible. So the only option is to derive a function from the desired traits of the currency.

Yes, I repeat stuff sometimes. That is because I forget I posted it before, so I post it again. You search on my posts and you are like, "Hey he posted that stupid story 3 times in the last 4 years." Of course, this is not necessarily bad, because let's face it, the average person is pretty stupid and you need to repeat stuff a lot before it sinks in.

Price stability is not a non-issue, it is a central goal of any monetary system. Gold achieves that goal because as it becomes more valuable, people work harder to extract it. For this reason gold has maintained a relatively stable value over the last 2,000. Believe or not the price of a goat, 2 gold dinars (or something like that, I forget exactly what it is), has remained pretty much the same as what it was in Mohammed's time (500AD).

For BC this will NOT be the case because there is an absolute limit to the supply. So, the policies of the existing BC client break one of the fundamental goals of a monetary system.

Also, although you can certainly start quoting things in micro Bitcoins or whatever, if you have any experience with accounting or invoicing systems you know this sort of thing is highly annoying and undesirable. Just waving your hands and saying a bunch of developers will solve this problem is something only a person inexperienced in these matters would say. If you have worked with real currencies and real retail systems you know the practicality of the denominational system is very important and has a significant impact on the willingness of people to use a currency.

[BitterTea]

Despite the current trend of Bitcoins losing value against the dollar I think long term the greater annoyance is going to be a constant rise in value. As more crypto currency is used its value will increase and since the quantity of coins is limited at 21 million the value of each coin will keep increasing. This will be annoying for two reasons:

The effective quantity of Bitcoins is not limited to 21 million. Loans increase the effective quantity of Bitcoins. Alternative blockchains increase the effective quantity of Bitcoins. It won't be a problem -- unless you imagine a powerful government that prohibits all business practices the increase the effective monetary supply, the effective monetary supply will grow as the economy does.

For example, imagine we're in a bitcoin-based economy. I want to buy a car that costs 5,000 bitcoins, but I don't have 5,000 bitcoins. I borrow 5,000 bitcoins and in exchange give an IOU worth about 5,000 bitcoins. The 5,000 bitcoins I borrowed are still in the economy, being spent by the guy I got the car from. But the IOU is also in the economy, and it effectively adds an additional 5,000 bitcoins because it is interchangeable in the marketplace with 5,000 bitcoins.

I still think you're wrong about this. Imagine Alice, Bob, and Charlie make up the entire economy. Alice wants the widget that Bob is selling for 2 BTC, but has none. So she borrows from Charlie and pays Bob. Bob now has 2 BTC and Charlie as an IOU from Alice for 2 BTC. Would Bob trade his 2 BTC to Charlie for Alice's IOU? Are they of equal value? Only if Bob trusts that Alice will repay 2 BTC.

The IOU is not fungible because its value depends upon the probability that the debtor will repay their debt.

[memvola]

Also, although you can certainly start quoting things in micro Bitcoins or whatever, if you have any experience with accounting or invoicing systems you know this sort of thing is highly annoying and undesirable. Just waving your hands and saying a bunch of developers will solve this problem is something only a person inexperienced in these matters would say. If you have worked with real currencies and real retail systems you know the practicality of the denominational system is very important and has a significant impact on the willingness of people to use a currency.

I started developing accounting systems twenty years ago in COBOL (yeah I was a late adopter, hehe). I've been staying away from such sins for at least ten years now, so I'm a little rusty, but what I can say is that designing a currency system to fit current accounting applications is remarkably backwards. Congratulations.

Price stability is not a non-issue

It's not, but it's not about deflation either.

gold has maintained a relatively stable value over the last 2,000. Believe or not the price of a goat, 2 gold dinars (or something like that, I forget exactly what it is), has remained pretty much the same as what it was in Mohammed's time (500AD).

If this is true, it's just coincidence. There is no fundamental reason this should be the case.

the policies of the existing BC client break one of the fundamental goals of a monetary system.

Nope.

[iddo]

For example, imagine we're in a bitcoin-based economy. I want to buy a car that costs 5,000 bitcoins, but I don't have 5,000 bitcoins. I borrow 5,000 bitcoins and in exchange give an IOU worth about 5,000 bitcoins. The 5,000 bitcoins I borrowed are still in the economy, being spent by the guy I got the car from. But the IOU is also in the economy, and it effectively adds an additional 5,000 bitcoins because it is interchangeable in the marketplace with 5,000 bitcoins.

Yes, you can say that the 5,000 IOU adds additional 5,000 bitcoins into the economy, but this IOU should be considered less valuable than actual 5,000 bitcoins because there's counterparty risk attached to it, i.e. the borrower might default in case he fails to earn 5,000 actual bitcoins to repay the loan.

Another issue that springs to mind: how could there be interest on loans if the money supply is fixed, i.e. where would the new money for the interest come from...? There's a simple example of how it would work at http://mises.org/daily/4569:

Suppose there are two neighbors living in a community that uses actual gold coins as its money. Mr. Smith starts out with 1,000 gold coins, while his neighbor Mr. Brown starts out with none.

Brown asks Smith if he can borrow 100 gold coins for a month. Smith agrees, but insists on a 12.7 percent APR, which works out to 1 percent interest charged monthly. So on day one of this deal, Smith has 900 gold coins while Brown has 100.

At the end of the month, after whatever buying and selling he has done with the rest of the community, Brown has accumulated 100 gold coins in his possession. He pays that back to Smith. Ah, but Brown is still short 1 gold coin. So he offers to cut Smith's lawn and paint his fence (with materials provided by Smith). In return for this labor, Smith pays Brown 1 gold coin, which Brown then uses to pay off his interest.

[DeathAndTaxes]

Also, although you can certainly start quoting things in micro Bitcoins or whatever, if you have any experience with accounting or invoicing systems you know this sort of thing is highly annoying and undesirable. Just waving your hands and saying a bunch of developers will solve this problem is something only a person inexperienced in these matters would say. If you have worked with real currencies and real retail systems you know the practicality of the denominational system is very important and has a significant impact on the willingness of people to use a currency.

Come on these are trivial problems.  It is a display issue nothing more.  You could make a modified client today which users enter payments in satoshis and it simply divides that entry by 1E-8 to get "whole BTC" behind the scenes. 

Only an inexperienced developer would think it is an issue.  Hell if you wanted to future proof your accounting software.  Record everything internally as integers in satoshis.  Then handle all I/O in another module which parses data in a variety of formats (BTC, mBTC, satoshis, etc).

[kgo]

Also, although you can certainly start quoting things in micro Bitcoins or whatever, if you have any experience with accounting or invoicing systems you know this sort of thing is highly annoying and undesirable. Just waving your hands and saying a bunch of developers will solve this problem is something only a person inexperienced in these matters would say. If you have worked with real currencies and real retail systems you know the practicality of the denominational system is very important and has a significant impact on the willingness of people to use a currency.

Come on these are trivial problems.  It is a display issue nothing more.  You could make a modified client today which users enter payments in satoshis and it simply divides that entry by 1E-8 to get "whole BTC" behind the scenes. 

Only an inexperienced developer would think it is an issue.  Hell if you wanted to future proof your accounting software.  Record everything internally as integers in satoshis.  Then handle all I/O in another module which parses data in a variety of formats (BTC, mBTC, satoshis, etc).

Actaully, I believe internally bitcoins are represented as 64 bit ints.  The current code actually adds the decimal place and translates behind the scenes.  Internally they are stored as Satoishis.

[DeathAndTaxes]

I thought that but I took a look at block explorer and it was in fractional BTC even when looking at raw view (maybe their "raw view" isn't that raw).  So I wasn't sure but I think you may be right.  I do recall reading something about internally the system handling everything as satoshis.

[BitterTea]

I thought that but I took a look at block explorer and it was in fractional BTC even when looking at raw view (maybe their "raw view" isn't that raw).  So I wasn't sure but I think you may be right.  I do recall reading something about internally the system handling everything as satoshis.

Yeah, a value of 1 = 1 satoshi; a value of 100000000 = 1 BTC.

[asdf]

Price stability is not a non-issue, it is a central goal of any monetary system.

The central goal of a monetary system is to provide a measure of value to facilitate trade. What's with this fetish for price stability? Stable against what? Ultimately, all prices are relative and floating against one another.

Here's a riddle for you: If corn halves in value relative to cotton, how much money do you print for price stability?

This is a total non-issue.

[stryker]

(1) the prices of things will be constantly changing since the currency will keep deflating

That's true of the US Dollar as well, but instead of the money losing purchasing power it will gain purchasing power. The rate of change will be driven lower by speculators long term (if everyone knows Bitcoins will rise in price, speculators will buy them, rising the price in the present). Over time, you will see a gradual, relatively predictable price and its annoyance for repricing goods should not be overwhelming. How many times per day does Wal Mart change the price on a good in their store? It's no great burden.

(2) we will be dealing in decimal units. Quoting ridiculous prices like BC0.000013 or whatever will be very annoying and a lot of invoicing systems are not designed to handle cents values below 1/100

Have some foresight!  A name will be given to smaller units. Perhaps the .00000001 unit will be called the Satoshi (seems to be catching on). In such a case where BTC is extremely valuable, your coffee will simply be "5.30 satoshis".   Nomenclature is a non-issue, the market will easily figure out efficient naming schemes.

I think ultimately we will have to make a competing client that breaks the 21 million limit and uses a more practical value scale and rate of inflation.

Go for it. I won't be buying it.  

if a satoshi is the smallest unit, how can a coffee be 5.30 satoshis? if they are the smallest unit then they are not divisible....

[memvola]

if a satoshi is the smallest unit, how can a coffee be 5.30 satoshis? if they are the smallest unit then they are not divisible....

Nice catch. That means precision will need to be increased before Bitcoin gains in value million-fold.

[DeathAndTaxes]

if a satoshi is the smallest unit, how can a coffee be 5.30 satoshis? if they are the smallest unit then they are not divisible....

Nice catch. That means precision will need to be increased before Bitcoin gains in value million-fold.

Exactly and the bitcoin network can handle any level of precision.  It currently is set at 1E-8.

At 1 BTC = $1,000,000 USD, 1 satoshi = $0.01 (at current limit)
Given the need for single penny units isn't exactly necessary we likely could get to 1 BTC = $10 mil and thus 1 satoshi = $0.10 before the current divisibility becomes an issue.

1 BTC = $10 mil is a monetary supply of $210 trillion and is roughly 5x the current global monetary supply.

These are trivial non-issue.  Broader acceptance, reduced volatility, more advanced tools, user friendly client, idiot proof backup & security these are much harder problems and have real impact on the uptake of Bitcoin.

[JoelKatz]

I still think you're wrong about this. Imagine Alice, Bob, and Charlie make up the entire economy. Alice wants the widget that Bob is selling for 2 BTC, but has none. So she borrows from Charlie and pays Bob. Bob now has 2 BTC and Charlie as an IOU from Alice for 2 BTC. Would Bob trade his 2 BTC to Charlie for Alice's IOU? Are they of equal value? Only if Bob trusts that Alice will repay 2 BTC.

For the original loan to take place, the IOU would have to be worth 2 BTC. That may mean its face value is substantially more than 2 BTC. That may mean it's insured.

The IOU is not fungible because its value depends upon the probability that the debtor will repay their debt.

The IOU might have a different risk/reward profile than bitcoins. But the risk/reward profile of currency is very hard to adjust. Meanwhile, those who craft IOUs can carefully tune their risk/reward profiles by choosing the face value, interest rate, and using partial or complete insurance. So it's quite likely that at least some people will prefer the IOU. And if you don't, you just have to sell it to someone who does. If, for example, the IOU's interest rate scales with inflation, it can actually have *less* risk than holding currency.

[JoelKatz]

Yes, you can say that the 5,000 IOU adds additional 5,000 bitcoins into the economy, but this IOU should be considered less valuable than actual 5,000 bitcoins because there's counterparty risk attached to it, i.e. the borrower might default in case he fails to earn 5,000 actual bitcoins to repay the loan.

Who said the face value of the loan was 5,000 bitcoins? I said the loan was *worth* 5,000 bitcoins (because a disinterested party paid 5,000 bitcoins for it in an open market). The value of a loan includes the risk.

Another issue that springs to mind: how could there be interest on loans if the money supply is fixed ...

The money supply is not fixed. Think about dollars for a minute. Bitcoins are like dollars but the Federal reserve can't print any more of them. So what? What percentage of US dollars are in the form of physical currency?

[DeathAndTaxes]

Your definition of money supply and the standard accepted definition of money supply are not the same thing.

Debt isn't part of money supply.  Debt is part of the larger economy but not the money suppy.
Debt isn't fungible, easily divisible, or have low friction.

The bitcoin monetary supply is fixed.
The US monetary supply is expanding.

Neither of those definitions require printed money.  The federal reserve doesn't print any money.  They expand or contract money supply via open market policies.

Just because something has an equivelent value in BTC doesn't mean it is part of the BTC money supply.

You have 500 BTC, I have one ounce of gold.  Using your debt analogy if I sell you the gold coin the BTC money supply has doubled because their is 500 BTC and 500BTC (equivelent) of gold.  In reality there isn't.  The money supply is still 500 BTC. 

I hate to quote wikipedia but:

In economics, the money supply or money stock, is the total amount of money available in an economy at a specific time.[1] There are several ways to define "money," but standard measures usually include currency in circulation and demand deposits (depositors' easily accessed assets on the books of financial institutions).

[Blinken]

While it is certainly true you can increase the supply of bitcoins by loaning them, even this is subject to the limitations of requiring a fractional reserve. Even the most reputable loaner will require a reserve to honor ongoing redemptions. Let's say, for the sake of argument the reserve is 10% and that 50% of the total supply is net loaned then perhaps we can expand the total supply by 500%. So, now, instead of 21 million coins we effectively have 100 million floating around. Once again the limit is fixed.

By loaning bitcoins we can expand the BC supply, not increase it indefinitely.

Posters arguing that price stability is not an important monetary objective need to go back to "skool" and take Econ 101.

Internally the bitcoin source code uses coins and cents as the fundamental units and formats and parses strings that way. Any change of denomination would require changes to the core logic of the current client.

[memvola]

Posters arguing that price stability is not an important monetary objective need to go back to "skool" and take Econ 101.

You can't enforce stability merely by a generation function, hence it's not relevant to the topic. If you have an idea about how to stabilize exchange value without compromising freedom from central authority, I'd like to hear it. At this point, I consider it impossible, since exchange value is external to the currency. In the same manner, you can't stabilize the value of corn relative to acetone without taking external measures either.

Internally the bitcoin source code uses coins and cents as the fundamental units and formats and parses strings that way. Any change of denomination would require changes to the core logic of the current client.

Cents? Correct me if I'm wrong but Bitcoin keeps it as a signed 64-bit Integer (1 BTC is 10^8 units). Increasing the precision is only possible with a change in protocol, but it can be seamless if done in a long interval, say two years. After that, clients older than two years wouldn't be able to connect to the network. No practical problems there. This is not "core logic" either.

[Etlase2]

You can't enforce stability merely by a generation function, hence it's not relevant to the topic. If you have an idea about how to stabilize exchange value without compromising freedom from central authority, I'd like to hear it. At this point, I consider it impossible, since exchange value is external to the currency. In the same manner, you can't stabilize the value of corn relative to acetone without taking external measures either.

Not to sound like a broken record in this thread, but check my sig

[JoelKatz]

I hate to quote wikipedia but:

In economics, the money supply or money stock, is the total amount of money available in an economy at a specific time.[1] There are several ways to define "money," but standard measures usually include currency in circulation and demand deposits (depositors' easily accessed assets on the books of financial institutions).

This is a reasonable definition of money supply. The fact that the block chain can never contain more than 21 million bitcoins puts no limit whatsoever on how many can be in circulation (just as a bearer bond can circulate the same way dollar bills do) or in demand deposits. The reason you don't see equivalents in circulation with dollars is because of a combination of legal restrictions and the lack of any need for them.

[DeathAndTaxes]

I hate to quote wikipedia but:

In economics, the money supply or money stock, is the total amount of money available in an economy at a specific time.[1] There are several ways to define "money," but standard measures usually include currency in circulation and demand deposits (depositors' easily accessed assets on the books of financial institutions).

This is a reasonable definition of money supply. The fact that the block chain can never contain more than 21 million bitcoins puts no limit whatsoever on how many can be in circulation (just as a bearer bond can circulate the same way dollar bills do) or in demand deposits. The reason you don't see equivalents in circulation with dollars is because of a combination of legal restrictions and the lack of any need for them.

It may be "reasonable" but it isn't used.  A bearer bond isn't part of US money supply.

For example no commonly accepted definition of US money supply includes mortgages which you would include for bitcoin economy.  Similarly the US money supply would be many magnitudes larger if you considered all mortgages, students loans, credit card balances, corporate bonds, derivatives, stocks, options, credit default swap, US national debt, etc to be part of the money supply.   It may be part of the economic activity but it isn't the money supply.

The same applies for global money supply and bitcoin money supply.

Debt isn't used to conduct trade.  Sure debt trades hands but it is the item being sold not the medium for trade.


People buy debt.  
People sell debt.  
However the other side of that trade normally involves money.

People buy potatoes
People sell potatoes
However the other side of that trade normally involves money.

Unless your definition of the bitcoin money supply includes potatoes it shouldn't include debt either.

Nobody for example goes to Grocery stores and pays for groceries with some debt they own.  The money supply is what facilitates trade not what is purchased.

Grocery <-> Currency
Debts <-> Currency
Videogames <-> Currency
...
Houses <-> Currency


The reason we track & measure the money supply is it is the intermediary that facilitates trade.  No matter what someone wants (from debt to videogames) they can acquire it via money and likewise when someone wants to sell something they can sell it for money.

If you trade debt for something other than money that is barter.  It doesn't make the debt part of the money supply.  Likewise if I trade you $1600 in potatoes for an ounce of gold are those potatoes part of the gold supply now.  Did the gold supply grow by $1600?  Of course not and likewise if you barter some debt you own for some other asset it doesn't increase the money supply.  I mean if you include debt in the money supply to not be arbitrary anything bartered for gold should be part of gold supply.

Lastly one trait of money is fungibility.  Each dollar is worth exactly the same as every other dollar.  Each unit of debt isn't worth the same (even if it has the same face value and interest).  Some will default, some will payoff early, some will payoff in full.  When you get paid in dollars (or bitcoins) it doesn't matter which dollar you get paid with however if you get paid in debt it does matter which debt you get paid with.  

While you can barter with debt it isn't money and thus isn't part of the money supply (any money supply).

[JoelKatz]

While you can barter with debt it isn't money and thus isn't part of the money supply (any money supply).

If that were true, then no fiat currency would be part of the money supply, inlcuding dollar bills. Fiat currency notes are debt -- most of the money supply for most of the world's nations are notes that represent sovereign debt. See, for example, 12 USC 411. http://www.law.cornell.edu/uscode/12/411.html

[DeathAndTaxes]

While you can barter with debt it isn't money and thus isn't part of the money supply (any money supply).

If that were true, then no fiat currency would be part of the money supply, inlcuding dollar bills. Fiat currency notes are debt -- most of the money supply for most of the world's nations are notes that represent sovereign debt. See, for example, 12 USC 411. http://www.law.cornell.edu/uscode/12/411.html

We will have to agree to disagree.

The money supply in many countries is based on debt but the actual money supply (as in the count) doesn't include debt.

Banks use fractional reserve banking to increase or decrease the money supply and central banks influence that action via monetary policy.

Still the ACTUAL MONEY SUPPLY = MONEY not debt.

As an example, the federal reserve defines the US money supply as:

M0: The total of all physical currency, plus accounts at the central bank that can be exchanged for physical currency.
M1: The total of all physical currency part of bank reserves + the amount in demand accounts ("checking" or "current" accounts).
M2: M1 + most savings accounts, money market accounts, retail money market mutual funds, and small denomination time deposits (certificates of deposit of under $100,000).
M3: M2 + all other CDs (large time deposits, institutional money market mutual fund balances), deposits of eurodollars and repurchase agreements.

Currently the M2 is ~10 trillion.  Obviously debt is not a component of the money supply given US mortgages and national debt are ~7T and ~14T respectively and collectively the US has roughly ~30T in debt of all forms.

M0, M1, M2, M3 are the measures of money supply and include ... money. 

The whole purpose of counting money supply is because the supply of money (not debt) affects prices.  People buy and sell things with money not fractional mortgage notes or partial credit card bonds.  Thus the amount of mortgages changing isn't going to directly affect the change in prices (inflation).  The change in the supply of money (actual money both physical and electronic) relative to the change in productivity is what results in inflationary or deflationary pressures.

[Steve]

While you can barter with debt it isn't money and thus isn't part of the money supply (any money supply).

If that were true, then no fiat currency would be part of the money supply, inlcuding dollar bills. Fiat currency notes are debt -- most of the money supply for most of the world's nations are notes that represent sovereign debt. See, for example, 12 USC 411. http://www.law.cornell.edu/uscode/12/411.html

In the US, FRNs are not debt, but they are backed by debt (on the FED's balance sheet)...that's an important distinction.  A large part, but not all, of that debt backing is US Treasury debt (in particular, the FED has added large amounts of MBS debt in recent years to its balance sheet).  However, when you deposit dollar bills into a bank checking account, that is debt (a debt the bank owes you).  And since the bank only maintains a fractional reserve backing of that debt and a checking account balance is for the most part as good as a physical note (especially due to FDIC insurance), that debt does effectively expand the money supply (of course there are different measures of money supply..."money supply" itself isn't a very specific concept).

[FreeTrade]

(1) the prices of things will be constantly changing since the currency will keep deflating

Just set your prices relative to something else you find desirable (or perhaps your input costs) and let software do the updating for you.

Quite Right!

May I suggest the USD is a rather stable unit of account. It is widely used and people are familiar and comfortable with it. This problem is largely solved.

I don't know why one would wish to use Bitcoin to solve a problem that is already solved, and for which Bitcoin is manifestly unsuitable, when there are so many unsolved problems for which Bitcoin is ideal.

[Steve]

(1) the prices of things will be constantly changing since the currency will keep deflating

Just set your prices relative to something else you find desirable (or perhaps your input costs) and let software do the updating for you.

Quite Right!

May I suggest the USD is a rather stable unit of account. It is widely used and people are familiar and comfortable with it. This problem is largely solved.

I don't know why one would wish to use Bitcoin to solve a problem that is already solved, and for which Bitcoin is manifestly unsuitable, when there are so many unsolved problems for which Bitcoin is ideal.

I'd rather based it on actual costs of other goods to remain independent of any other currency.  For curiosity, back in July I started tracking the prices of 9 staple items at my local grocery store.  The prices were 2% higher at the last check about a week or so ago.  That's a 2% rise in ~3 months!

[JoelKatz]

Banks use fractional reserve banking to increase or decrease the money supply and central banks influence that action via monetary policy.

Still the ACTUAL MONEY SUPPLY = MONEY not debt.

Fractional reserve banking is debt. My $100 checking account balance is the bank's debt. If a checking account balance is part of the money supply, then debt can be part of the money supply.

Your dollar bill is debt. It is an obligation of the United States. If a dollar bill is part of the money supply, then debt can be part of the money supply.

Whether or not debt actually *will* be part of the money supply depends on a lot of factors. The primary one is whether or not it makes economic sense for it to be. In a hypothetical economy with a currency shortage, there will be a demand for currency, and debt will meet that demand.

[FreeTrade]

I'd rather based it on actual costs of other goods to remain independent of any other currency.  For curiosity, back in July I started tracking the prices of 9 staple items at my local grocery store.  The prices were 2% higher at the last check about a week or so ago.  That's a 2% rise in ~3 months!

Again, most people will probably find it convenient to use the USD together with a measure of inflation when calculating costs (assuming US based). Finding an accurate figure to approximate the inflation in the basket of goods you typically buy is tricky, but doing the calculations yourself would give you an exactitude that is probably not worth the candle. (your figure of 8% p.a. for staple goods sounds about right to me)

[Etlase2]

Death, you have a misunderstanding of fractional reserve.

If I deposit 1 million BTC in a bank, and they have a 10% reserve, they can loan out 900k of that BTC. I still have 1 million BTC. 1.9 million BTC now exists from that 1 million.
This is the exact same thing that happens with fiat.

Since there can be no FDIC in bitcoin, people will probably not want to bank with a bank that only keeps a 10% reserve because if there is a bank run, they are forked. But 10% reserve banks will have the highest interest (or however it works in a deflationary economy), so you will have greedy people that will use them. But hey, that's their problem if they can't get paid when the bank goes bankrupt.

[Steve]

If I deposit 1 million BTC in a bank, and they have a 10% reserve, they can loan out 900k of that BTC. I still have 1 million BTC. 1.9 million BTC now exists from that 1 million.

That's incorrect...if you deposit 1 million BTC in the bank and they operate on a 10% reserve policy, they can then lend up to 9 million additional BTC.  They would have claims of 10 million BTC in their customers' accounts and 1 million backing those claims (10%).  Of course they're not creating BTC out of thin air, they are actually only just updating an account balance (or they could issue their own "Debt Coins fractionally backed by BTC").

[Etlase2]

Right, I was just pointing out the first step in the chain.

[Steve]

Interestingly, many countries have much lower than 10% reserve requirements, and some have no reserve requirements at all.  In a system that doesn't have FDIC like insurance, where banks are allowed to fail, a bank can be left to decide for themselves what level of reserves they maintain (but buyer beware).  A bank could conceivably make loans and issue a currency (account balances) without any backing at all.  They just have to convince people that it has some value in some way.  They could setup an exchange that trades their currency for something else of value and establish some level of support without any explicit reserve backing.  Over time, people could come to trust the value of that currency and the bank could gradually withdraw its support from the exchange.  As long as the bank didn't inflate the currency, it could continue to maintain value.  It's not that dissimilar to bitcoin except that you have to trust this central issuer.

[Blinken]

As I already pointed out a bank cannot have a 0% reserve because in that case depositers would not be able to withdraw cash. You always have to maintain enough cash to cover fluctuations in depositer withdrawals. Also, any bank will have instruments, like CDs, constantly coming due. You have to have cash reserves to pay these debts.

[JoelKatz]

Since there can be no FDIC in bitcoin, people will probably not want to bank with a bank that only keeps a 10% reserve because if there is a bank run, they are forked.

They won't be able to get their money immediately, but a bank can easily prevent a bank run from being particularly disastrous. What they have to do is explain in advance what their bank run policy is and have sufficient assets of some kind (which can be loans) to cover the run in the long term.

For example, if the prevailing interest rate is 4%, they can have an "emergency run interest" of 9%. If there's a run on the bank, they declare an emergency and cover withdrawals with 9%, fixed-term notes. Sure, some people won't be too happy to have these notes instead of bitcoins, but their value will likely be a bit higher then the corresponding number of bitcoins, so they could likely sell them for a slight gain.

This assumes the bank isn't in danger of going out of business. This means the loans (or other bank assets) must still be sound. This can only cover a short-term liquidity run problem, not a long-term solvency crisis. (It wouldn't have helped with the bank failures we saw recently.)

However, a bank can easily cover a bank run with non-cash assets. This is another way commodities can be used to inflate the money supply. You might not trust a bank to keep only a 10% bitcoin reserve, but if they had a billion dollars worth of gold or also owned many stable businesses, you probably would. So a bank can leverage any other asset to allow them to put more bitcoins into circulation than they could otherwise.