Freidman K Percent Rule

[spartacusrex]

Hello,

I am brain storming about a coin that doesn't have a fixed limit on it's production..

I am wondering what techniques could be used to find a vale to use for the Monetary Expansion of a coin.

Freidman had a very simple rule, that he believed was better than leaving this to the discretion of central banks. A rule that was set in stone.

Judging by recent events, I'm not sure central banks can be trusted, or their discretion considered, well.. Good.

If you launched a coin, that had a monetary supply growth policy, what options are there other than 'pick a number', say 4%, and grow at that ?

Basically, getting information, about the outside world, into the chain is difficult.. maybe impossible.

I was thinking that maybe you could use the price of a fixed coin, say Bitcoins, in this new coin, as a marker for how much to expand the money supply..

Maybe the price of Gold as a factor ? (if you could somehow tell the chain..)

Any ideas ?

Is the K percent rule OK or has it been debunked.. ?

My analysis of the situation says that actually, K percent rule works well in most cases..

http://economics.no/Friedman.pdf

http://pages.uoregon.edu/gevans/SJPE_05005004.PDF

[1714994573]

1714994573

[1714994573]

1714994573

[ElectricMucus]

Just tie the monetary base to the computational power. Currency is added or removed based on the difference between current and previous "difficulty".
Of course that requires some working incentive to provide computational power except currency generation.

If this is possible however I think that would be the ideal case.

[spartacusrex]

That's a very interesting idea..

I like it because it doesn't require any external knowledge, off chain..

But - does that mean if the Hashing power increased by a factor of 10, say asics become the norm, that you should increase the monetary supply by x10 ?

Probably not.. some factor of Hashing power could be used..

Growing the money supply at the rate of economic growth, would seem the ideal, so the questions are I suppose :

Is there a link between Hashing power and Economic growth ?

What other factors might this effect.. ?

Is it worth risking this as opposed to a fixed k percent ? (Since this might have quite large uncertainty..)

[ElectricMucus]

That's a very interesting idea..

I like it because it doesn't require any external knowledge, off chain..

But - does that mean if the Hashing power increased by a factor of 10, say asics become the norm, that you should increase the monetary supply by x10 ?

Probably not.. some factor of Hashing power could be used..

Well, yes and no. I said computational power, that is on it's own much more than solving a particular problem. Obviously we have to take a compromise here and define the resources that a eligible to be worth money.
Consider FFT for instance, it is used in pretty much any application and it can be used for cryptographic purposes too.
 

Growing the money supply at the rate of economic growth, would seem the ideal, so the questions are I suppose :

Is there a link between Hashing power and Economic growth ?

What other factors might this effect.. ?

Is it worth risking this as opposed to a fixed k percent ? (Since this might have quite large uncertainty..)

This is a complicated problem.
The size of an economy should be somewhat related to the available computational power, with a catch....
In any case we wouldn't be able to measure the whole magnitude, only the amount dedicated to the system, which results in an interesting problem:

The percentage of computational power which can be wasted is proportional to economic growth. So during a recession period the monetary base would be inflated despite the need to do otherwise, resulting in well inflation beyond economic growth and so currency devaluation. I think the crux of the matter is how useful the cryptographic work can potentially be, in case of Bitcoin it is useless, but I don't think it is necessary that proof of work should consist of wasted resources.
For instance we could compromise the granularity of the currency base and come up with a way to make only certain transactions possible which correspond to the result of useful computational work. It isn't really necessary to have a system where you can send any amount of currency down to a certain precision. For example one would only be able to choose out of certain predetermined amounts.
Every address would have certain amounts it could send to any other address, which corresponds to a certain fraction of the available funds.
In detail if an address currently has 10 credits, there could be the option to send 23/1234, 83/991,... of the available balance to one adress and  53/256, 3/10, ... to another address.

A more down-to earth alternative would be some sort of contest where currency would be allocated to whoever comes up with a solution to a real problem first, despite several people working on the same stuff, efficiency wouldn't be that high but it's better than nothing.

If all else fails and proof of work really requires truly wasted resources we can at least choose one which is based on a complicated enough mechanism which can be useful if otherwise needed, like FFT.