Network-wide self-corecting mechanisms

[HZPyR8eVk]

Consider including in the BitCoin protocol some network-wide self-correcting mechanisms. By this I mean that the BitCoin P2P network participants would have to be able to change some of the BitCoin protocol parameters, in their client application.

The only clear idea that I have right now is the maximum amount of BitCoin in circulation. Not necessarily the absolute amount, but as an annual inflation percentage. For example, the maximum amount for 2010 is 21 million, for the next year could be 1% more, and so on for each year (cumulated, so as to simulate system / population growth, but not as a percentage value, just as style).

This percentage can be either integrated in the protocol, or left as a BitCoin client editable parameter, in which case the final value can be settle upon through agreement among the vast majority of BitCoin clients.

Reasons for this type of inflation: to avoid deflation and creation of credit.

Some people suggest that there could be a credit market in the BitCoin market. But consider that mining gold (and fiat printing) has been replaced with individual BitCoin mining, and thusly eliminated the centralized and authoritarian injection of money in the economy (through gold fields claimed by states, gold mining corporations and central banks).

BitCoin should not aspire to replace money (what is money: gold, fiat?), but aspire to replace what works as money (= the money supply, which includes credit).

I mean, why would Bitcoin try to replace just a small part of the centralized and authoritarian injection of money in the economy, when the credit market accounts for far more things which work as money?


George Hara

[1714062483]

1714062483

[Quantumplation]

The 21 million is a hard cap that we'll eventually reach.  The number of bitcoins in circulation for 2010 is (roughly) 2620800 coins.

Each block generates 50 coins.
In 2 weeks, 2016 blocks (should) be generated.
In a year, there are 26 "2weeks".
50*2016*26 = 2620800

The idea of increasing the supply of money relative to time span, population, transactions, etc. has  been discussed at length.

[kiba]

BitCoin should not aspire to replace money (what is money: gold, fiat?), but aspire to replace what works as money (= the money supply, which includes credit).

Money is whatever people agree to use as a medium of exchange. Thus, bitcoins are money.

[HZPyR8eVk]

BitCoin should not aspire to replace money (what is money: gold, fiat?), but aspire to replace what works as money (= the money supply, which includes credit).

Money is whatever people agree to use as a medium of exchange. Thus, bitcoins are money.

The problem is that credit constitutes the vast majority of the (fiat) money supply, not cache. BitCoin should not be overwhelmed by the same issue (because that credit is a centralized monetary mechanism).

[FreeMoney]

 

  (because that credit is a centralized monetary mechanism).

I promise you it isn't. I can extend credit in any currency I have to anyone I choose.

It may seem this way because the money you happen to use is printed by one special organization, but that is not a property of all money. And even with central money I can extend credit up to the amount I have.

[HZPyR8eVk]

  (because that credit is a centralized monetary mechanism).

I promise you it isn't. I can extend credit in any currency I have to anyone I choose.

It may seem this way because the money you happen to use is printed by one special organization, but that is not a property of all money. And even with central money I can extend credit up to the amount I have.

I don't understand what you mean?! You can have all sorts of credit (like with your neighbor), but what does that buy for most people? The vast majority of the world's money supply is created by monopolistic state law, which in turn means by monopolistic credit entities.

For BitCoin it makes no sense to have a few entities generating credit out of thin air, when BitCoins can be directly generated out of thin air by individuals (well, it's indirectly from electricity). However, such entities would start popping up and create credit if the BitCoin supply remains limited.

I mean, then people could just use Ripple and be done with the whole concept of medium of exchange.

[Red]

Consider including in the BitCoin protocol some network-wide self-correcting mechanisms. By this I mean that the BitCoin P2P network participants would have to be able to change some of the BitCoin protocol parameters, in their client application.

The only clear idea that I have right now is the maximum amount of BitCoin in circulation.

In most cases, a single person can't change bitcoin parameters unilaterally. That's sort of like saying I use gold for coins, but you choose to use silver. But you want to pretend that both of our coins are the same.

However, everyone could change to a different set of parameters at once if necessary. However, there are reasons each of the parameters was set at its current value. I just don't know what those are.

So I'm going to post that question in a new thread.

[kiba]

For BitCoin it makes no sense to have a few entities generating credit out of thin air, when BitCoins can be directly generated out of thin air by individuals (well, it's indirectly from electricity). However, such entities would start popping up and create credit if the BitCoin supply remains limited.

I mean, then people could just use Ripple and be done with the whole concept of medium of exchange.

I am not sure about what you mean. It's impossible to have a central credit or central bank in the bitcoin economy, unless we're talking about a very good bank that crushed the competition.

[joechip]

BitCoin should not aspire to replace money (what is money: gold, fiat?), but aspire to replace what works as money (= the money supply, which includes credit).

Money is whatever people agree to use as a medium of exchange. Thus, bitcoins are money.

The problem is that credit constitutes the vast majority of the (fiat) money supply, not cache. BitCoin should not be overwhelmed by the same issue (because that credit is a centralized monetary mechanism).

And with the correction going on right now... most people call it 'recession' or 'depression' or 'slow-down' credit is being de-valued.  What is being destroyed right now is credit, but not money.  The AMB and M1 are rising, meaning the Fed has increased the size of the base money supply it is doing so to reflate the credit markets, but those markets are not needed right now, hence the value of them is falling and the price of commodities is rising in relation to its supply. [1]  Therefore, the commodity value of money is rising due to a liquidation of the malinvestment brought on by the manipulation of interest rates, which coordinate capital needs (commodities) through time.

The failing of many of the current crop of deflationists (Prechter, Mish, Denninger) is that they have this quaint notion that credit money is indistinguishable from commodity money [2] and the destruction of it means the destruction of commodity prices along with it.  They are wrong.  They are two distinct types of money and one can rise while the other falls.  This sends mixed pricing signals in the market and breeds all kinds of confusion, especially amongst economists, most of whom are morons.

Ta,

[1] - by holding steady, ie. the CCI is in a tight trading range, commodities are rising in value vs. the amount of credit available. 
[2] - in a fiat system, the monetary base, as defined by the central bank's balance sheet acts as the 'commodity' money.

[HZPyR8eVk]

For BitCoin it makes no sense to have a few entities generating credit out of thin air, when BitCoins can be directly generated out of thin air by individuals (well, it's indirectly from electricity). However, such entities would start popping up and create credit if the BitCoin supply remains limited.

I mean, then people could just use Ripple and be done with the whole concept of medium of exchange.

I am not sure about what you mean. It's impossible to have a central credit or central bank in the bitcoin economy, unless we're talking about a very good bank that crushed the competition.

It is virtually impossible to have a central bank in BitCoin, but it is possible to have credit creationists (when the amount of money in circulation is fixed and the economy is growing; why? because there is demand for stuff which acts as money / medium of exchange), and while there would be competition, they would be significantly fewer (that is, concentrating more power over the BitCoin economy) than the total number of BitCoin peers.

Why let those few increase the money supply through credit (and thusly concentrate the power to increase the money supply), when everybody can do the same from their own BitCoin application?

[HZPyR8eVk]

BitCoin should not aspire to replace money (what is money: gold, fiat?), but aspire to replace what works as money (= the money supply, which includes credit).

Money is whatever people agree to use as a medium of exchange. Thus, bitcoins are money.

The problem is that credit constitutes the vast majority of the (fiat) money supply, not cache. BitCoin should not be overwhelmed by the same issue (because that credit is a centralized monetary mechanism).

What is being destroyed right now is credit, but not money.  The AMB and M1 are rising, meaning the Fed has increased the size of the base money supply it is doing so to reflate the credit markets, but those markets are not needed right now, hence the value of them is falling and the price of commodities is rising in relation to its supply.

Right. The important point here is that it was the central bank (= Fed) who create the bubble by having those credit mechanisms, and now the market has to correct it.

So, why allow credit in the first place in BitCoin, when the cache itself can inflate as the economy's needs grow.

My point is, implement in BitCoin the mechanisms which allow the majority of peers to control the supply. (Not so that things would get out of control, but some leeway.)

[kiba]

Right. The important point here is that it was the central bank (= Fed) who create the bubble by having those credit mechanisms, and now the market has to correct it.

So, why allow credit in the first place in BitCoin, when the cache itself can inflate as the economy's needs grow.

My point is, implement in BitCoin the mechanisms which allow the majority of peers to control the supply. (Not so that things would get out of control, but some leeway.)

If the market want credits, they will get it. Beside, bitcoins is already decentralized. What more decentralization do you want?

What the hell are you asking?

[Gavin Andresen]

If the market want credits, they will get it. Beside, bitcoins is already decentralized. What more decentralization do you want?
What the hell are you asking?

He/she/they want an easier way for The Crowd to control The Policy-- right now, if you know enough C++ to modify and compile your own version of Bitcoin and can convince enough other people to use your version, then you can change Policy.

Allowing everybody to tweak Policy and having some sort of mechanism that figures out what "everybody" wants to do is an interesting idea.  I have no idea how you could actually make it work, and it would open up a whole other can of potential security problems (what if somebody controls a bunch of IP addresses and decides to "vote" for a policy that benefits them?  or can you tie the policy changes to proof-of-work somehow?  How do you aggregate what "everybody" thinks in a non-spoofable way? etc etc etc)...

[bytemaster]

This probably belongs in the economics section but we need to throughly squash the notion of falling prices being bad.   We need to fully establish the practice of "double lending" as a means of creating "money" is what is bad.  It is easy to see because the "double lender" gets something for nothing and thus the purchasing power of all of society is transferred to this "double lender".  When credit expansion fails to grow exponentially people default (not enough new money to pay interest on old).   This creates a chain reaction and people realize that banks leveraged their money 100 to 1 and so a 1% fall wipes it all out. 

Bit coin prevents fractional reserve banking!   Sure a bank could take bit coins as a deposit and then issue bit notes.  But the key is that everyone will know that the notes are a debt of the bank and not real money.  Notes were adopted because they were easier to use than real money (gold) and people accepted them "as gold".  With bit coin there is no reason to accept anything in place of a bit coin.  If you take a bit note you are really "trusting the bank" as you have not yet received your real coin.  In this digital world you would simply immediately cash your note.  Thus no one would "keep the notes in circulation" and any fractional reserve bank would immediately be caught in a "bank run". 

A currency represents a claim on goods.  He who can print / mine a currency transfers wealth from others.  So any "inflation" scheme that is not the equivalent of a "stock split" transfers wealth.  So, for everyone out there who says they want the CPI to be flat or even growing at a any rate you would simply multiply everyones account by the same "fixed number".   But what would be the practical difference for every day users?  If a gallon of milk is 1 BTC and deflates to .5 BTC and I have 0.5 BTC in my account I can buy a gallon of milk.  If you double the money in everyones account then milk will cost 1 BTC again but I can still only buy one gallon.  Flat prices no transfer of wealth.

Thus divisibility is the key with periodic "bit shifts" to rebase how the currency is displayed.  Purely cosmetic.  The key is NO WEALTH TRANSFER from the holders of money to the creators of money.

[HZPyR8eVk]

With bit coin there is no reason to accept anything in place of a bit coin.

Everything you said is clear to me, but I disagree on one point.

If there is deflation and people start hoarding money thinking that their money will value more later (so they have an incentive to start spending their money later - half the money for milk), credit creationists would intervene on this opportunity. It's a matter of timing. So, just because of this timing (a few entities are far more mobile than the entire BitCoin network), you end up having a few entities hoarding the power to create stuff which works as money, rather than allow BitCoin itself to do that directly for everybody.

People would accept this credit because its available and works as money. Consequently, there is transfer of wealth (from the entire network to the few credit creationists).

Might not become significant, I have no idea, but I gave this as an example for the ability of the network to correct itself.

[HZPyR8eVk]

He/she/they want an easier way for The Crowd to control The Policy

Exactly. Right now BitCoin is centralized in the sense that the rules are established by Satoshi, in code. Is it good to have a limit for the money supply? I have no ideea (I believe it's not because of what I said in the previous post: timing). Does Satoshi actually know that? I think the network itself would know better.

I have no idea how you could actually make it work, and it would open up a whole other can of potential security problems (what if somebody controls a bunch of IP addresses and decides to "vote" for a policy that benefits them?

Well, today if the majority of the network were controlled by rogue peers, they would in effect control BitCoin, so I don't see a difference. However, I don't know enough architecture of either BitCoin or P2P networks to say it can work effectively, with no more potential security issues.


Any peer editable parameter could be changed within certain limits, say once a year, and only if a high enough percentage from the peers change that parameter. Of course, not all peers would use the same value, some would want more, some less; just take an weighted average. Perhaps the per-peer voting power should be weighted by the total mining / spend throughput, or something. Don't make the change instant; wait a few months while things cool down and the peers change their protocol parameters.


I've tossed this idea around, maybe it can be useful to the people involved in designing the protocol. The money supply is just an example.

[Red]

Allowing everybody to tweak Policy and having some sort of mechanism that figures out what "everybody" wants to do is an interesting idea.  I have no idea how you could actually make it work, and it would open up a whole other can of potential security problems (what if somebody controls a bunch of IP addresses and decides to "vote" for a policy that benefits them?  or can you tie the policy changes to proof-of-work somehow?  How do you aggregate what "everybody" thinks in a non-spoofable way? etc etc etc)...

Votes could be weighted based upon your address' balance of bitcoins. Stock proxy-vote style. The rich would have more power.
Votes could be weighted based upon the number of nodes. The hackers would have more power.
Votes could be one-person, one-vote. But that would have to be verified outside the system.

[Red]

f there is deflation and people start hoarding money thinking that their money will value more later (so they have an incentive to start spending their money later - half the money for milk), credit creationists would intervene on this opportunity.

This logic is mathematically flawed. People don't realize it because they have never seen it.

If external value starts to flock to bitcoin, than the external value to trade increases, while the number of coins to trade stays the same. That means if the value of external commodities traded doubles overall, than the value of each bitcoin doubles as well.

Example: If there were 100 loaves of bread a day and 100 bitcoins the mean value of bread is 1 BTC. If you want to trade an additional 100 loaves a day, you either have to double the velocity of the coins, or trade in 1/2 BTC coins at the same velocity.

That's the easy part! What is counter intuitive is the dynamic vs lending.

In this situation your argument says that people would start lending their excess bitcoins (at interest I'm presuming). So what would the interest rate be in this situation? So if we presume that the amount of value people want to trade using bitcoins will double this year (not unlikely) what is a reasonable interest rate? Well, just putting them in my mattress will give me 100% interest in commodity value with 0% risk.

However, if I want to borrow the money to start a business manufacturing some commodity widgets, the if I borrow the equivalent of 100 widgets today in BTC, then I need to pay the BTC back with the equivalent value of >200 widgets in a year. So if my widget business is growing at 50% growth a year, I'm screwed.

Deflation make lending very expensive, and very risky. That is why there is minimal lending now, EVEN THOUGH the government is giving away free money for banks to lend.

People are not used to this in their regular life because they are used to price inflation.

Inflation means that money in your mattress is worth less to you (in commodities) next year than it is worth today. That encourages you to LOAN your money to a bank in hopes that they will pay you at least as much interest to make up the difference. A little extra is even better!

This makes lending cheap for new businesses and much less risky for lenders. Business need a much smaller growth rate and profit margin to succeed. Including the bank to whom you lent your money.

[bytemaster]

In a highly deflationary environment lenders would lend money at negative interest rates and still make a real profit. 

The real problem is that people like the idea of numbers getting bigger more than they like the idea of smaller numbers doing more.    Who wants to take a nominal pay cut! 

So to counter this mental handicap of society, it may make sense to simply debase the currency via a "stock split/rebasing" to keep a basket of goods a constant nominal BTC value.   In effect this would make the "real return on investment" caused by deflation tangible to the thick headed public that thinks 4 pennies is better than 1 nickel because 4 > 1. 


Assume the number of goods in the economy grows at 5% per year. 

Normally Joe earns 100 BTC this year and a gallon of milk costs 100 BTC.   The economy grows by 5% thus the price of milk would fall to 95 BTC; however, the nominal BTC value of Joes labor also fell and his boss gave him a 2% nominal pay cut and the average Joe gets ticked (too stupid to realize he really got a 3% raise). 

Now if you "rebase the currency" then the price of milk will remain 100 BTC, Joe's boss gives him a 3% raise and the average joe is "happy" even though his employer really gave him a real pay cut!  But because the price of milk stayed flat "joe" feels richer (and he IS richer because all of society has more goods), but not as relatively rich as he would have been under deflation.   

Where as if we did not do any "stock splits" Joes salary would have fallen to 98 BTC but milk would now cost 95 BTC. Clearly hie is richer.     

In my view it is a GREAT THING when someone loses their bit coins because the value (read purchasing power) of lost coins gets evenly distributed among all other coins. This is a WIN for all of society except the poor sap who lost his coins.

[Red]

In a highly deflationary environment lenders would lend money at negative interest rates and still make a real profit. 

This seems quite silly as opposed to not lending the money and making even more real profit.

Perhaps I don't understand what you are saying?