The hoarding problem

[nebulus]

. . . Since they are *basically infinitely divisible this is easy to accomidate . . .

The way computers manage numbers does not allow bitcoin to be infinitelly divisible . . .

Each bitcoin is currently divisible into 100,000,000 pieces.  That should be plenty for a long time.  Will we ever need to divide it further?  Impossible to predict, but if we do I feel certain that the modifications necessary will be possible.  It isn't going to happen for as long as it isn't necessary, and I'm probably going to be long gone and dead before it becomes necessary, so I'm not too concerned how they choose to handle it.

Bitcoin community can not just decide to divide further. For this new computers are needed or significant revamp of software. If you know anything about how computers works you know what a double is. The range for this data type is -9,007,199,254,740,992 to 9,007,199,254,740,992. Compare this to total number of allowed satoshis which 21,000,000,000,000,00 this is also 16 digits.  So you can say that the number of satoshi is maximized based on the modern day hardware. I am skeptical that the latter is going to change by much in the coming years there is simply no need for that in mainstream.

[1714810921]

1714810921

[1714810921]

1714810921

[1714810921]

1714810921

[DannyHamilton]

Bitcoin community can not just decide to divide further . . .

Don't be silly, of course they can.  How they'll choose to do it will depend on the options available if/when it is determined to be necessary.

I'm confident that when it becomes necessary (if it ever becomes necessary), the best method available at that time will be implemented.  I don't expect to be alive at that time, so I don't find it to be very important to me.

[wareen]

. . . Since they are *basically infinitely divisible this is easy to accomidate . . .

The way computers manage numbers does not allow bitcoin to be infinitelly divisible . . .

Each bitcoin is currently divisible into 100,000,000 pieces.  That should be plenty for a long time.  Will we ever need to divide it further?  Impossible to predict, but if we do I feel certain that the modifications necessary will be possible.  It isn't going to happen for as long as it isn't necessary, and I'm probably going to be long gone and dead before it becomes necessary, so I'm not too concerned how they choose to handle it.

Bitcoin community can not just decide to divide further. For this new computers are needed or significant revamp of software. If you know anything about how computers works you know what a double is. The range for this data type is -9,007,199,254,740,992 to 9,007,199,254,740,992. Compare this to total number of allowed satoshis which 21,000,000,000,000,00 this is also 16 digits.  So you can say that the number of satoshi is maximized based on the modern day hardware. I am skeptical that the latter is going to change by much in the coming years there is simply no need for that in mainstream.

I hope you're not suggesting to use a floating point data type for storing an amount of Bitcoins, because that's a really bad idea! Anyway, 64 bit fixed point data type allows for 3 more decimal places (2^64 > 2.1*10^18) so Bitcoin isn't maxing out the hardware - even if we do not change the width of the data type used in the current transaction format.

[Walter Rothbard]

I find this concept is hard to explain to people, but the price of bitcoins and the number of bitcoins is irrelevant. If say half of all bitcoins were lost suddenly then the market would adjust. Assuming the same demand, the remaining coins would simply be worth twice as much. Since they are *basically infinitely divisible this is easy to accomidate.
Compare it to gold again. If half the worlds gold were to disappear then it's scarcity would be higher and so would the price. This could be true even if there were so little gold that it were traded in atoms instead of ounces.    

RodeoX, you might like Murray Rothbard's explanation in the first part of his book What Has Government Done to Our Money?  He gets to the question in chapter 8.  It's not exactly easy reading up till that point, but the thought experiment he gives in that chapter is understandable enough to explain to your grandmother.

[DeathAndTaxes]

. . . Since they are *basically infinitely divisible this is easy to accomidate . . .

The way computers manage numbers does not allow bitcoin to be infinitelly divisible . . .

Each bitcoin is currently divisible into 100,000,000 pieces.  That should be plenty for a long time.  Will we ever need to divide it further?  Impossible to predict, but if we do I feel certain that the modifications necessary will be possible.  It isn't going to happen for as long as it isn't necessary, and I'm probably going to be long gone and dead before it becomes necessary, so I'm not too concerned how they choose to handle it.

Bitcoin community can not just decide to divide further. For this new computers are needed or significant revamp of software. If you know anything about how computers works you know what a double is. The range for this data type is -9,007,199,254,740,992 to 9,007,199,254,740,992. Compare this to total number of allowed satoshis which 21,000,000,000,000,00 this is also 16 digits.  So you can say that the number of satoshi is maximized based on the modern day hardware. I am skeptical that the latter is going to change by much in the coming years there is simply no need for that in mainstream.

Nothing about this is correct.  Nothing.  

Bitcoin uses "bigint" constructs to perform 256 bit math.  By your logic 256bit encryption is impossible on current computers because they natively don't support 256bit numbers.   It is done in software.  You can do 256bit math on an 8 bit computer.

Likewise Bitcoin uses ulong (64 bit unsigned integer) to support transaction values however it still works fine on 32 bit (and 16 bit, and in theory 8 bit) computers.  The idea that more advanced computers or some radical redesign is required is simply wrong.

Bitcoin could be hard forked to use a 128 bit integer to represent transaction values and thus up to 31 digits of precision without any radical change in code or hardware requirements.  There is absolutely no reason to do so but there is no reason to be making up stuff.

[BTCdude007]

I believe the "hoarding" problem will resolve itself as the price rises. As people have more and more capital gains on their bitcoins, they will be tempted to cash out and spend. Also, as the bitcoin "market cap" rises, more businesses will start accepting BTC to get a piece of the action.

[nebulus]

. . . Since they are *basically infinitely divisible this is easy to accomidate . . .

The way computers manage numbers does not allow bitcoin to be infinitelly divisible . . .

Each bitcoin is currently divisible into 100,000,000 pieces.  That should be plenty for a long time.  Will we ever need to divide it further?  Impossible to predict, but if we do I feel certain that the modifications necessary will be possible.  It isn't going to happen for as long as it isn't necessary, and I'm probably going to be long gone and dead before it becomes necessary, so I'm not too concerned how they choose to handle it.

Bitcoin community can not just decide to divide further. For this new computers are needed or significant revamp of software. If you know anything about how computers works you know what a double is. The range for this data type is -9,007,199,254,740,992 to 9,007,199,254,740,992. Compare this to total number of allowed satoshis which 21,000,000,000,000,00 this is also 16 digits.  So you can say that the number of satoshi is maximized based on the modern day hardware. I am skeptical that the latter is going to change by much in the coming years there is simply no need for that in mainstream.

Nothing about this is correct.  Nothing.  

Bitcoin uses "bigint" constructs to perform 256 bit math.  By your logic 256bit encryption is impossible on current computers because they natively don't support 256bit numbers.   It is done in software.  You can do 256bit math on an 8 bit computer.

Likewise Bitcoin uses ulong (64 bit unsigned integer) to support transaction values however it still works fine on 32 bit (and 16 bit, and in theory 8 bit) computers.  The idea that more advanced computers or some radical redesign is required is simply wrong.

Bitcoin could be hard forked to use a 128 bit integer to represent transaction values and thus up to 31 digits of precision without any radical change in code or hardware requirements.  There is absolutely no reason to do so but there is no reason to be making up stuff.

64 bit unsigned long integer is exactly 16 digits, same as double (I confused the internals with what users sees).

Whether 128-bit hardware can be done in software or not is just a speculation until demonstrated. Until then I consider it impossible. I read somewhere that to fill a 128-bit memory (or filesystem) would require more energy than boiling the oceans. Can not find the original article.

Now, I do not know what kind of effect on Bitcoin there would be if amounts are done in structs to expand the precision range. I do know it would be a hell of a job for a programmer to do because that would require storing one amount in multiple memory locations. It is possible, but is it practical?

It is also curious that the limit Satoshi chose is tied up to mainstream CPU/memory addressing. Maybe the answer does not lie in the fact that he wanted to limit the supply, maybe the answer lies in the fact that it is impossible?

[DeathAndTaxes]

Whether 128-bit hardware can be done in software or not is just a speculation until demonstrated.

No it just means you are uniformed.  128-bit, 256-bit, 512-bit (any number of bit) math can be done in software regardless of the hardware native register size.  Computations at the register size can be done faster and more efficiently but any sized values can be operated on using any sized CPU.  Bitcoind for example runs on a 32 bit machine (despite using 64bit ints for values and 256bit ints for some cryptographic functions). 

Bignum - This library performs arithmetic operations on integers of arbitrary size. It was written for use in public key cryptography, such as RSA and Diffie-Hellman.
http://www.openssl.org/docs/crypto/bn.html

[bitfarmer]

Why is this referred to as a "hoarding" problem? Why is saving not the correct term for this? To me, it's derogatory to call saving hording. At some point, the coins will be spent or sold. Hoarding is what those people on TV do, keep it until it is trash. Unless it's your plan to keep your BTC until you format them away, it's not hoarding.

[nebulus]

Whether 128-bit hardware can be done in software or not is just a speculation until demonstrated.

Bignum - This library performs arithmetic operations on integers of arbitrary size. It was written for use in public key cryptography, such as RSA and Diffie-Hellman.
http://www.openssl.org/docs/crypto/bn.html

cool, thanks...

[sounds]

Why is this referred to as a "hoarding" problem? Why is saving not the correct term for this? To me, it's derogatory to call saving hording. At some point, the coins will be spent or sold. Hoarding is what those people on TV do, keep it until it is trash. Unless it's your plan to keep your BTC until you format them away, it's not hoarding.

Correct.

Because bitcoin is a deflationary currency it will be attacked by all who believe such currencies are doomed.

Attackers (debaters if you prefer) should use logic to discuss the relative merits of fiat currency versus bitcoin. However, they are likely to also try emotional attacks, and calling bitcoin users "hoarders" is an emotional attack. It isn't logical, but that wasn't the point of the attack.

[Lethn]

Nearly all of the conventional currencies in our countries worldwide dramatically punish these 'hoarders' you speak of who are actually just saving to spend or for emergencies, how exactly is this a bad thing? I keep trying to wrap my head around how people are accusing Bitcoin users of 'hoarding' when really it's just saving up for the most part so they can afford expensive things, that's how you do things in normal life. You get a loan from a bank in order to pay for it in 60 years and then just dump the problem on your children which is what everyone is doing now.

I actually looked around at prices again recently and I just realised because of Bitcoin being a deflationary currency I would actually stand a chance of being able to afford car insurance for myself in the year if I earned something! I'd have to work a bit hard for it still because the prices are outrageously high but I'll at least be able to afford it now because of the currencies' purchasing power compared to the pound. I'd also like to point out that the system of printing money also causes dramatic price rises and falls as we are seeing constantly around us so if this 'hoarding' ( lol I still can't get over the fact someone used that term to describe a deflationary currency ) really does become a problem and I doubt it will at least people will be able to prepare for it somehow. You do forget though that there is such a thing as a luxury market, some people out there may well be saving up like crazy so they can afford things like that, this is how you get rich people to part with there money.

I suspect you have been reading/listening to a neo-keynesian economists, I really do think these guys are wankers I'm sorry because they cause frightening amounts of destruction to emerging and established markets and they think they're doing the work of God, it's incredibly easy to see just by looking at history how badly they are wrong, even Marx had better logic than these guys.

[odolvlobo]

Why is this referred to as a "hoarding" problem? Why is saving not the correct term for this? To me, it's derogatory to call saving hording. At some point, the coins will be spent or sold. Hoarding is what those people on TV do, keep it until it is trash. Unless it's your plan to keep your BTC until you format them away, it's not hoarding.

Both "hoarding" and "investing" are forms of saving. A person that has savings can either "invest" it, and return it to the economy, or "hoard" it and remove it from the economy. It is a common term in economics and not meant to be derogatory (unless you are a Keynesian).

I would also like to point out that deflation by itself is not a problem at low levels. Just like you can have hyper-inflation where nobody wants to hold the currency, you can have hyper-deflation where nobody wants to spend the currency. Both are economic disasters. However, hoarding and deflation do not automatically lead to hyper-deflation, just as borrowing and inflation do not automatically lead to hyper-inflation.

[RodeoX]

P.S. I added the " * " to the word basically to avoid the math details. I do understand it cannot be truly divided infinitely, but enough to accommodate a lot of money.   

[Walter Rothbard]

Why is this referred to as a "hoarding" problem? Why is saving not the correct term for this? To me, it's derogatory to call saving hording. At some point, the coins will be spent or sold. Hoarding is what those people on TV do, keep it until it is trash. Unless it's your plan to keep your BTC until you format them away, it's not hoarding.

Yep.  The real problem is that some people would like to own more of the money of others, so they denigrate the decision of others to save by calling it "hoarding."  There is no mathematical delineation between hoarding and saving.  It's just a matter of choosing a disrespectful term to try to force the outcome you want, and using that disrespectful term to try to apply public pressure.  There's no real such thing as "hoarding."

[Walter Rothbard]

Why is this referred to as a "hoarding" problem? Why is saving not the correct term for this? To me, it's derogatory to call saving hording. At some point, the coins will be spent or sold. Hoarding is what those people on TV do, keep it until it is trash. Unless it's your plan to keep your BTC until you format them away, it's not hoarding.

Correct.

Because bitcoin is a deflationary currency it will be attacked by all who believe such currencies are doomed.

Attackers (debaters if you prefer) should use logic to discuss the relative merits of fiat currency versus bitcoin. However, they are likely to also try emotional attacks, and calling bitcoin users "hoarders" is an emotional attack. It isn't logical, but that wasn't the point of the attack.

Exactly!

And the truth is, if people don't like bitcoin's limited supply, they can simply choose to not use it.  The fact that they argue it usually points to the fact that they want input into the decision of others, not just themselves.

[deadweasel]

Why is this referred to as a "hoarding" problem? Why is saving not the correct term for this? To me, it's derogatory to call saving hording. At some point, the coins will be spent or sold. Hoarding is what those people on TV do, keep it until it is trash. Unless it's your plan to keep your BTC until you format them away, it's not hoarding.

Yep.  The real problem is that some people would like to own more of the money of others, so they denigrate the decision of others to save by calling it "hoarding."  There is no mathematical delineation between hoarding and saving.  It's just a matter of choosing a disrespectful term to try to force the outcome you want, and using that disrespectful term to try to apply public pressure.  There's no real such thing as "hoarding."

This is exactly right.  Word choice is crucial when judging anything.  It's so hard for anyone to be objective without strong mindfulness.

[twolifeinexile]

Any Turing complete machine can do what ever other Turing complete machine could do, (given enough memory)

I think divisibility of bitcoin would not be any problem in the long run.

Whether 128-bit hardware can be done in software or not is just a speculation until demonstrated.

No it just means you are uniformed.  128-bit, 256-bit, 512-bit (any number of bit) math can be done in software regardless of the hardware native register size.  Computations at the register size can be done faster and more efficiently but any sized values can be operated on using any sized CPU.  Bitcoind for example runs on a 32 bit machine (despite using 64bit ints for values and 256bit ints for some cryptographic functions). 

Bignum - This library performs arithmetic operations on integers of arbitrary size. It was written for use in public key cryptography, such as RSA and Diffie-Hellman.
http://www.openssl.org/docs/crypto/bn.html

[twolifeinexile]

Ultimately, limited supply nature distinguish bitcoin from other fiat/electronic currencies,this is one key feature of bitcoin, and probably why it attracts people. (Too much inflationary, central controlled currencies already)

The question in my original post is if the "saving" habits together with the deflationary aspects of bitcoin cause its overall impact on economy limited in scope. I fully expect bitcoin to be successful as a special "asset", but as a "currency", if it is constantly appreciating against other everyday life assets like housing/food/energy, then if people have choices of token asset to spend (either a fiat currency, a bitcoin, a central controlled electronic currency, a whatever medium of exchange with certain popularity), people will choose the one has high probability of depreciating in value to spend. The supplier, on the other hand, is in the position earning money, so usually want to accept what ever that is popular/liquid in nature. ( He can then later exchange this popular currency into a currency that is deflationary to "save" or "hoard" ). By the mechanism of these micro-economic behavior, deflationary currency have a tendency to quit the market ( as a currency, not as an asset)

Why is this referred to as a "hoarding" problem? Why is saving not the correct term for this? To me, it's derogatory to call saving hording. At some point, the coins will be spent or sold. Hoarding is what those people on TV do, keep it until it is trash. Unless it's your plan to keep your BTC until you format them away, it's not hoarding.

Correct.

Because bitcoin is a deflationary currency it will be attacked by all who believe such currencies are doomed.

Attackers (debaters if you prefer) should use logic to discuss the relative merits of fiat currency versus bitcoin. However, they are likely to also try emotional attacks, and calling bitcoin users "hoarders" is an emotional attack. It isn't logical, but that wasn't the point of the attack.

Exactly!

And the truth is, if people don't like bitcoin's limited supply, they can simply choose to not use it.  The fact that they argue it usually points to the fact that they want input into the decision of others, not just themselves.

[dancupid]

I currently sell items on eBay and bitmit. The items I sell on bitmit are about 0.6 the price of the items I sell on eBay.
ie I've made an economic decision that bitcoins are more attractive than paypal +fees.

People with bitcoins can expect cheaper prices, and they can decide based on the current situation whether to use Bitcoin or use fiat.
The hoarding argument assumes that the bitcoin price will just be taken from the current mtgox $ price directly.
But a merchant who wants Bitcoins will price their product accordingly.

Market forces will work this out naturally.