Topic: A Better Coin - page 2. (Read 12445 times)

Let me try to explain.

Transferring mining rights to wealthholders is how you would reduce the costs of security. In theory, I think it is the only possible method.

One way of thinking about it is as follows:

There are two ways to secure the network:
1) Require people to spend resources to vote on block validity
2) Screen people who vote on block validity to ensure that they are trustworthy.

People who are heavily invested in the system do not have incentives to carry out attacks. Successful attacks would undermine the value of their assets. Would someone who already holds 50% of the total coin want to carry out double spends? No, that person would be a fool to do so. Accordingly, he is rather trustworthy as a miner. If you subsidize wealth-holders in their search for blocks (for example by allowing them to submit blocks even with a low difficulty threshold), then you are screening miners for trustworthiness. Miners who are not trustworthy (people without a large stake in the system) would have to spend more computing resources to mine the same amount of blocks.

By shifting the security technology from resource use to voter screening, you save on resource costs and make network security cheaper. Ultimately, this means you can lower txn fees substantially without endangering the network.

Yes, I remember that idea, but I have to admit I don't understand it.
If it is feasible, of course reducing the costs of security would be great.


I know, for you is just as bad as inflation and you only want the inflation to reward miners perpetually.
But if you think rewarding miners perpetually is necessary, don't you prefer demurrage over inflation?


We could study deeper the usefulness of the internal variables for stable prices, but you don't seem very interested on that topic.
I think that studying the velocity looks more promising than difficulty. Difficulty provide information of the ming market instead of the economy as a whole. On the other hand, velocity could be manipulated (at a lost with mandatory minimum fees).


No I don't thank that mining rights should be transferred to wealthholders. But if they could collaborate for free (or fairly rewarded) in the security of the network I would not oppose.

I did not want to make computing power vote, remember? You are kind of assigning a platform to me that I never advocated. I want wealth-holding to be the main determinant of whether you are able to find a block and consign computing power to a marginal role. Do you agree that it would be better to make mining ability based on 90% wealth-holding and 10% computing power rather than 100% on computing power (as it is now)?

I don't think that preset rules for demurrage are a good idea. I prefer the current protocol. If there is no voting, I think the only sensible change is to link currency generation/destruction rates to difficulty growth. If there is voting, other alternatives would also make sense. Voting based on wealth requires reform of the mining process. Thus, the only possible channel of agreement for us is the mining process. Do you think mining rights should be transferred to wealthholders? I haven't really heard any response on this.

We could debate the details of mining reform, but the debate would be pointless unless everyone (or at least someone besides me) agrees that it is desirable first.

I think we're getting close to the consensus that demurrage is better than inflation.
Many of us agree we don't want direct human intervention in the system, too.
Until you remove money holders voting from your proposal I don't think it can be accepted by many people.
Only miners can vote and, as you pointed out, the effects of their voting must be innocuous for them or they will vote in their favor.
So the idea of miners reporting exchange prices is also banishing.
I still think they can vote to define a parallel stable currency for contracts and prices if it doesn't affect anyone's balance.

We don't seem to be getting anywhere. I view my suggestions as improvements vis-a-vis the bitcoin status quo. I am highly skeptical of suggestions made by everyone else.
Have we come to any kind of consensus about anything yet? Has anything been proposed that everyone thinks is a good idea or has the makings of a good idea?
If not, is there any point in continuing the discussion?

Exactly, that's the key for my point 2 here.
Differences between inflation and demurrage:

1) The nominal. The accountability inconvenience that inflation produces is not needed. I'll bring an exponential growth example in the hope that you start to dislike it.

5% infaltion rate
year 0, 1 coke = 1 inflatacoin
year 1, 1 coke = 1.05 ic
year 15 1.05^15 = 2.07892818 ic
year 23 1.05^23 = 3.07152376 ic
year 29 1.05^29 = 4.1161356 ic
year 33 1.05^33 = 5.00318854 ic
year 37 1.05^37 = 6.08140694 ic
year 40 1.05^40 = 7.03998871 ic
year 43 1.05^43 = 8.14966693 ic
year 45 1.05^45 = 8.98500779 ic
year 47 1.05^47 = 9.90597109 ic
year 49 1.05^49 = 10.9213331 ic
year 51 1.05^51 = 12.0407698 ic
year 43 1.05^53 = 13.2749487 ic
...
see? the time to gain just one unit is getting shorter and shorter until eventually will be less than a year, then less than a mounth, less than a day, less than an hour, less than a second...

2) The financial

Inflation rises nominal interest and leaves real interest untouched. Demurrage, on the other hand lowers real interest and encourages investments.
This seems in contradiction with what we see today, but today inflation is not created by mining but is loaned to existed at low interest. This represents a transfer from the conventional lender (saver) to the privileged borrower (the government and commercial banks). [Well, it also represent transfer of wealth from every money holder to those same privileged borrowers].
The financial market becomes unfair and allows investments that are not based on saving, with catastrophic malinvestment effects.
But let's just compare our far less dangerous mining inflation with demurrage in financial terms.

A) 5% inflation, 5% interest
Since profits tend to zero by competition, the capital yield of our example tends to be equal to the interest rate, 5%.
Why the interest rate is a lower limit to capital yields?  
Well, if someone borrows at 5% to invest in a 4% capital, his investment is not solvent. The only way to not loss money there is to sell the capital at a bubble price.
So our bakery costs 10000 at 5%.
His bakery yields 5% of its value annually, but the price of the capital is also rising.
In the second year, the 5% yield of its inflated 10500 will be higher than his agreed 500 annual interest quote and the difference are profits for the baker.
After 15 years, he can sell the inflated bakery to pay the principal of the loan and take the difference in price for himself.
Because the lender knows this and he's the one who decides if the bakery is built or not, he's able to charge inflation premium and the baker will still be able to pay him back (but the baker will have no profits).
When negotiating, the lender says: you pay me 10% interest rate or I keep my money (or invest it the money myself). I won't lose anything and you lose a job as baker.
So the nominal interest rate in this case would be 10% = 5% real/basic interest (or liquidity premium) + 5% inflation premium.

B) 5% demurrage, 5% interest

Here, when negotiating the lender cannot argue that he can keep the money, because he cannot deny that he will do it at a lost.
I know, with inflation the money holder is also losing but not in the same way. While the effect of demurrage is automatic and equal to every money holder, the monetary inflation precedes the price inflation. The miner has to spend his new money to create the price inflation. First he will inflate a certain market (whatever he demands) but that imbalance will be transferred to other sectors when the receiver spends it again. It will take some time until an equilibrium is reached again and this new money to affect the whole economy equally.
So let's say that the baker is brave enough to ask for a 4.9% interest rate. The money holder thinks:
"With inflation I had two options: lending at a 5% real interest rate or investing for a 5% yield, but now I can lend at 4.9% or invest it myself for 5%"
He says to the baker, "You know what? I'll buy you the bakery. Your rent will be 500, because that's what the bakery can yield."
But many lenders have done the same and the greater competition in the bakery market leads the yield of bakeries to drop to 4.9%.
Next time, when the baker asks for a 4.8% instead of 4.9%, maybe the lender says: "ok, just take it".
The yield of the bakeries will drop until all that you can make by renting one until the time destroys it equals the costs of building it, until the profits of building a bakery are zero.
Interest impedes capital profits to drop to zero.

But anyway, if you don't believe all this freeconomist theory, you still should chose demurrage over inflation.
Why not chose the alternative that is better for accounting purposes?
Has inflation any advantage over demurrage?


Yes. e-gold had a demurrage of 1%. They could have defined a parallel unit without demurrage and falling conversion rate, but the effect is the same: you will receive 1% less gold next year.

Except with demurrage the debasement happens instantaneously, and evenly across currency owners.  With inflation, prices gradually get bid up as the new money works its way through the economy.  I imagine this can be potentially disruptive, as the new money starts overly concentrated in one place.  My intuition tells me bubbles could be caused this way.

Maybe I'm swaying back to the demurrage side now...

If it's a backed, centrally issued currency, would a reduction in the bank's conversion rate achieve the same thing?
EDIT: I think it would, provided it's able to use variable conversion fees effectively to thwart any speculative attacks on it.

I've been beginning to lean this way too, after proposing my idea here with demurrage.

My position on this is that the psychological benefit, if any, is with inflation, not demurrage. Inflation is an every day phenomenon. Everyone is familiar with txn fees and inflation. Few people have ever seen a government drain their bank accounts to keep prices stable. I would rather not impose an unfamiliar experience on end users. I think this idea will be poorly received  by 99% of the population of potential users.

Nope.

If I have 14T coins, and the only way for someone to get one is to buy a coin from me for $1, but I guarantee to buy it back for $1, then all I have to do is ensure I have enough USD stored to cover the outstanding coins.

I question the legality of it though.  With it being centralized, I could imagine the government could (and would) shut it down pretty quickly.

In that case, I would prefer demurrage because it keeps prices stable.  Even though it is a purely psychological benefit, it's better than no benefit achieved with inflation!

However, I'd still like to see a deflation rate roughly equal to GDP growth, again, with the purpose of keeping prices stable.

There is no real difference. You are reasoning correctly. The only distinction is a purely nominal one. Unless psychological illusions are introduced, demurrage and inflation produce identical real outcomes.
In theory, perfectly predictable demurrage yields real outcomes 100% identical to those under perfectly predictable inflation at an equivalent rate.


Demurrage:

Monetary outcome: I have 1 coin.  If I store my coin for one year, it erodes to 0.5 coins.   Price of a coke is constant over time at 0.5 coins.
Real Outcome: I choose between purchasing two cokes now or one coke one year from now.  

Inflation:

Monetary Outcome: I have 1 coin. If I store my coin for one year, I still have 1 coin. Price of coke increases from 0.5 coins now to one coin one year from now.
Real Outcome: I choose between purchasing two cokes now or one coke one year from now.  

Notice that the real outcomes are identical. If the real decision you make (drinking a coke now or drinking a coke one year from now) remains the same, then the two monetary systems are equivalent in real terms.


You might ask why there is so much fuss about demurrage then? I think the simple answer is that the discussants are not well trained in economics. Confusion abounds and ignorance reigns.  


There is a lot of focus on money velocity here. In Keynesian economics, both increases in inflation and increases in demurrage have positive affects on velocity. People want to sell their coin for goods before value erodes. However, as you can see in the Coke example, the real effects of inflation and demurrage in this respect are 100% equivalent.

It requires an initial capitalization of 14 trillion USD to be fully sustainable.

No, demurrage just keeps the velocity constant as people always have the same urge to spend their money quickly to avoid losing it. In our inflation equation it means %ΔV equals 0 or very close to it, so it is now a known variable mostly under our control. Demurrage is a great to mitigate deflation as people will keep spending and investing even if their purchasing power is going up (unless the deflation rate surpasses the demurrage rate, then hoarding becomes more profitable).

This is what I read to get it:
http://forum.webofdebt.com/viewtopic.php?f=8&t=231
http://www.arbitragemagazine.com/features/miracle-town-worgl/
http://en.wikipedia.org/wiki/Demurrage_(currency)

After that I reread jtimon's posts about it.

Since you brought it up JohnDoe, I still haven't been able to really figure out the difference between demurrage and inflation.  If we had a 5% demurrage rate, but 4% inflation, wouldn't any GDP growth above 1% cause prices to rise?

Oh, and here's a new thought I had.  What about DollarCoin?  Someone starts a blockchain with a 14T coin genesis block, and no other coins would be generated.  A very small demurrage rate would ensue on any coins NOT in the original genesis block (say, 0.285%, to make up for lost coins).  The person managing the blockchain would guarantee to buy and sell the coins for $1 a piece.  So it would be completely and entirely backed by the dollar.

Aside from being centralized, does anyone see a problem with this?  Seems like a good way to create a completely stable currency.  Well, stable compared to the dollar anyway.

I finally understand the effects of demurrage and think it's a very good policy to incorporate.

I think that the currently proposed annual demurrage of 0.285% won't be enough to deter hoarding during times of deflation though. How about a 5% demurrage which goes to miners instead of being destroyed? Aside from keeping the velocity fairly constant this would completely eliminate the need for the "ramp up" period and immediate reward expansion as this would basically be a tax on early adopters that's given back to late adopters through block rewards. We would no longer have to go out of our way to solve the early adopter problem so we could start immediately with 500k/block rewards and maintain it like that until inflation goes down to our target 4%. This would greatly reduce the number of years we remain in high inflation.

Also a higher demurrage could help with initial business adoption as it makes people more eager to spend the currency.

Effects on miners are complicated. The key element here is that miners' earnings are proportional to transactions volumes.

Basically:

If people purchase more real goods and services, then miners earn more in real terms
If people purchase fewer real goods and services, then miners earn less in real terms.

Unless you know how inflation/destruction decisions affect the real volume of purchases, it is very difficult to figure out what the implications for miners are. My best guess is that miners would want to maximize the  market cap of the currency. This is only a guess though.

The more important issue here is whether mining should be changed to make computing power less important. In my opinion, It would be better if a mix of computing power and wealth was used to mine rather than just computing power. This would greatly improve blockchain security. I'm pretty sure this is feasible, but we should agree on whether this is desirable before debating practical solutions.

Part of the reason why I favor mining reform is because I think bitcoin should be like a joint-stock company. Decisions in a joint-stock company can be made on a voting basis (one share = one vote). Shareholders try to make decisions that benefit the company. Allowing computing power to vote is not a good idea. It is like assigning votes to people who don't own stock. If mining were reorganized so that wealth played a primary role, then mining-based voting could be used to make decisions.

Don't worry.


What I mean is that everyone will have the same purchasing power (only with more or less coins) but the miners.
I meant the generation reward, not the fees.
If they earn 50 coins but a lot of coins are being created in everybody's account, the earn more (in purchasing power).
If they earn 50 coins but a lot of coins are being destroyed in everybody's account, the earn less (in purchasing power).

But thinking it better, they earn the same because the prices will adjust fast.
If miners do that voting they will probably vote for destruction and deflation, because they don't get the inflation coins (the holders do).
I still don't think that money holders voting is very practical.

Sorry about being an arrogant prick. I should have explained it better to begin with. I often blame others for my poor communications skills (personality defect).

Yes, you are getting it right. Yes, I mean proportionally.

There is considerable variation in the miners' reward due to changes in velocity. Bitcoin will also have this property once currency generation stops.
A new twist in this system is that velocity affects the rate at which the money supply can change.

Low velocity -> blocks with few txns -> miners are paid few rewards -> blocks can generate or destroy few coins.
High velocity -> blocks with many txns -> miners are paid many rewards -> blocks can generate or destroy many coins.  

Assuming people vote for price stability (the likely outcome I think):
If a lot of coins are being spent and price is falling, you have a lot of tax revenue to destroy. Destruction puts upward pressure on price.
If a lot of coins are being spent and price is rising, you have a lot of tax revenue to distribute to coin holders. Distribution puts downward pressure on price.

If few coins are being spent, you have very little control over price.

Now you've explained it better. When you say "evenly across all existing balances" I assume you mean proportionally, because then you say "Given that your share of the pie is independent of your vote...".
Here the only notable effect is the variation in the reward for the miners (which stays nominally constant but changes its "share of the pie"). Am I getting this right?


Please don't assume that you're smarter than me only because I disagree with you. It's a very dogmatic and pretentious attitude.
As far as I can tell, I made always logical constructs when refuting other people's arguments.