Bitcoin Forum

hi everybody!

i've been reading here for a while and by now i think i understand most of the aspects of bitcoin. but i really don't get, why the total amount of BTC is limited to 21M? as far as i know, there is no limited fiat currency out there!

if bitcoin refuses to fail and gets more and more accepted a limited amount means ever increasing deflation which rewards hoarding over spending?! sounds more like a bug than a feature to me. at least in the long run!

xx
viajero

I think one theory is that a shrinking Bitcoin economy will also cause the value per Bitcoin to decrease (prices for goods increase), which would give hoarders an incentive to spend at least part of their savings. As the economy grows, the value increases (prices for goods decrease), giving a bigger incentive to save. There's a balance between saving and spending which an economy will find naturally.

i don't see, how it would "balance out naturally"!
either the bitcoin economy grows and we have ever increasing deflation or it decreases and BTC will fail eventually.

even without growing we'll have a slight deflation due to every day losses (hardware failure, lost private keys, etc) that happens to every currency and thats why central banks print money every day!

i understand, that mining for ever is problematic too.

but what i'm actually asking myself is this: has the limited amount of BTC any strong theoretical reason/explanation or is it only because there is no better solution for a decentraliced (crypto-) currency?

Bitcoin's only method of money supply contraction is through destroyed coins, which are proportional to holdings. Holding coins effectively carries a risk and storage fee, so people will not have incentive to hoard.

The exact number of 21M is arbitrary and could be any number.

The limited supply mimicks gold which has worked as a money for human societies for many years. Economists disagree whether limited supply and deflation is good for society - on one side you have the neoclassics, keynesians, and the monetarists that believe some inflation creates more economic growth, on the other hand you have the Austrian school that argues market participants will anticipate inflation and it therefore doesn't help. They also argue that deflation doesn't hurt.

sure, as long as the costs and risks outweight the deflation, hoarding makes no sense. but providing appropriate technical know how, these risks and costs are pretty low. so the only significant risk is inflation which is not going to be expected, as long as BTC doesn't fail.

and isn't hoarding, growing acceptans, speculation, economic growth, etc some kind of money supply contraction too?!

If you think inflationary fiat is so much better, then just stick to that.

The topics you're mentioning have been discussed to death. Please do a little reading before rehashing something for the 10,000th time.

Welcome to the board.

If these risks are low, then money supply contraction will be low. If 2% of bitcoins are lost every year, you will, on average, lose 2% of bitcoins every year. There is no reason to hoard.

You should rationalize this away and not pay attention. I will not be replying.


https://bitcointalk.org/index.php?topic=158111.0

well, i did a lot of reading over the past months and actually jumped on the BTC train fortunately long before it reached the 10€ mark :) believe me, i'm a strong believer in BTC! i was just wondering about the reason why and the theoretical background of the 21M

i did some research for this topic as well, before i posted my initial post but couldn't find any satisfying results. well... i did some more research and found that:

https://bitcointalk.org/index.php?topic=11627.0 and that:
http://bitcoinweekly.com/articles/one-apple-today-two-apples-tomorrow-or-how-i-stopped-being-afraid-and-learned-to-love-deflation

it doesn't really answers all of my questions, but it clarified a lot.

There actually isn't one, it's an arbitrary number.  Actually, we come to 21 M because of how the block reward subsidy is issued.  50 BTC per block for the first 4 years (10.5 M BTC) and then it halves, and continues to half until it's too small to represent with a 64 bit integer.  No matter how long that takes, the total number of bitcoins issued will always approach, but never quite reach, 21 M BTC.

But that's actually inmaterial.  It could have been any number, as long as it had a maximum.  Bitcoins are infinately divisable under the protocol, and are presently divisable to eight decimal places.  It's really just a reference point.

it seems my subject is missleading.

"why is the total amount limited" would be more precise.


that's interesting! so the satoshi isn't the smallest possible piece? is it really infinately divisable?

"Discussion to death" is a measure of correctness in which logic system?  ::)

He failed to tell you why that is irrelevant to your theoretical question.

Probably he doesn't even have a clue what I mean.

No one know why, because the Creator(s) are not know.  However, a fix amount make scene,  because no one has control over the system, so who would make that decision to increase the coins.  Anyway Bitcoin is divisible to 8 places at the moment, so there is enough to go around

The system could make the decision mathematically.

I certainly do.  The real question is, "why is there a limit?"  The simple answer to that question is that Bitcoin was designed as a digital approximation of a finite natural resource, such as gold.  Actually, precisely as gold.  Every decision concerning an arbitrary metric within the bitcoin protocol was intentionally biased to approximate the mining of gold as best as was reasonablely possible.  My point was that the limit isn't a problem in the same manner that a limit on a fiat currency would be, because a BTC is (potentially) infinately divisable, so there cannot ever be a liquidity crisis as a result of a limited currency base.

Yes, but clients cannot use more than 8 decimal places because of a technical limitation.  Specificly, the value of any address is stored as a 64 bit integer.  All bitcoin addresses store only satoshis, not BTCs, and the clients  present the totaled values to the user with the decimal point added for human readability.  The decimal point doesn't actually exist in the current address structure, but the protocol permits new kinds of addresses to be created for any number of reasons.  The first ascii character of your bitcoin address denotes the address type, which is why all currently valid bitcoin addresses start with a "1".  The addressses designed for testing features on testnet are identical except that first character is different.  I believe that it's an "a" IIRC.  An address that used a floating point variable for the value could store sub-satoshi values.  Future address versions could also use different encryption algos than is currently used, or be different in a number of other ways.

I find your comment very interesting. I didn't know that the software only sees satoshis (0.00000001 BTC). But how could this be implemented? Is it feasible to do this? Or would it have to be in another altcoin?

8 decimal places seems to be a design flaw, as a 64-bit integer can store 11 decimal places at Bitcoin's current limits. Testnet addresses begin with 2.

Excepting that, as I mentioned, the current addresses don't store floating point variables, but integers instead.  So the decimal point for BTC are actually near the middle of a 64 bit integer for convience, and a single address can thus hold the entire 21 million BTC with room to spare and never create a buffer overflow error.  An address with a floating point variable would be able to store fractional satoshis, but wouldn't necessarily be able to store the entire monetary base.  This should never happen anyway, but it's more of a security/code issue than an economic one.

It's more than feasible, it's part of the future development plans of Bitcoin from nearly the start.  Maybe even the start.  As I said, Bitcoin was designed for future flexibility in addresss types, so that new address types can be created, and easily identified by both human and machine, without breaking backward compatibility with the current address type.  The future clients will just have to check the first ascii character to know how it should process that address; and teh running network won't even skip a beat.  There is much to the finer details of Bitcoin that are so elegant and so well thought out, I have long said that Satoshi was either a polymath or three or more geniuses in their own fields working in concert.

Perhaps it could, but how would that work?

Therein lies the rub, no one has been able to work out a better solution than what Satoshi has devised from the beginning.  His solution is both elegant, well interconnected, and easy to understand (from a theory basis, not from a practical implimentation basis).

If you decide that the number of bitcoins should grow indefinitely, the you have to decide what the rate is.

Ideally, you would want the rate to match economic growth so that the purchasing power of a bitcoin would remain constant, but I don't think that would be feasible at all. It's basically what we have now, and look what has happened.

So, you have to come up with a fixed rate. I don't think anyone can determine what the best fixed rate would be -- an arbitrary one must be chosen. I think the best rate is 0 because it makes all the math easier and more straightforward, and it eliminates some uncertainty.

It's not 21 million bitcoins, its 21 billion millibitcoins, 21 trillion microbitcoins. Its the way currency should be, as in not punishing you for keeping it safe. Keynesian dollars force you to risk currency just to break even... antithetical to the NAP if you ask me.

thx for all the helpfull answers :)  


honestly, for me the keynesian argument (not as a whole, but in this specific manner) totally makes sense. it forces the wealthy ones in a society to do more with their money than just sit on it. in a by methematical law deflationary currency the pressure to invest and therefore support the economy is much less, than in a inflationary one. the current crisis hasn't been caused by "keynesian dollars", it's caused by pulling down almost all the financial market restrictions and regulations. therefore we can thank our weak governments which are much more influenced by neoliberal lobbyists, than by elections or reason or anything else...

afaik there hasn't been anything compareable to bitcoin before, so we can only speculate what is going to happen. nobody knows for sure, the future will educate us ;)

It doesn't, because the wealthy are the only group that can efficiently save their funds in gold, land, etc.  It's much more difficult and economicly inefficient for the poor and middle classes to do the same thing.  Thus, inflation harms the lower classes more, percentagewise, because they have to keep a majority of their savings in dollar denominated assets for liquidity reasons.