Has anyone proposed a coin that adjusts its supply to stabilize prices?

[timhuge]

To stabilize the price against what?

The goal is to stabilize the prices in B2C (or whatever you want to call the units of this alternative coin) of the goods and services people buy with it. Of course, the relative prices of different goods change, so the goal of price stability is typically conceived in terms of a price index like the Consumer Price Index. What I'm proposing is not to use such an index directly in the code - I'm just saying that the success of my scheme should be measured against an ideal such index.

What I think is possible to implement with minor technical tweaks to Bitcoin is pegging the value of 1 B2C to the real cost of computing one hash. (This peg could change over time according to a pre-determined formula chosen to anticipate changes in the cost of computing, like Moore's law and energy prices.) To the extent that the real cost of computing a hash changes in unanticipated ways relative to the ideal price index, the price index for B2C will still exhibit price inflation or deflation. But that volatility should be much much less than what we are seeing with Bitcoin.

In essence, you are proposing to replace (to some extent) volatility in price with volatility in money supply. Both are prone to speculative games and manipulation.

Still, I like this thread, except that it lacks specific, detailed proposals about the idea you presented in broad strokes above.

<goes to work detailing specific proposal>

edit: To be clear, I'm at work and sort of booked the next couple days. But I'm planning on writing a little analysis up within the week.

[yvv]

I don't see how it could be done with a minor change specifically on the technical side??
Elaborate.

Well, the following could be done.

Кeep all ask and bid  orders for all fiat currencies versus BTC in the blockchain. This is possible without changing the protocol. Allow peers to vote for each order, just like they vote to confirm or reject the transaction. Orders which are going to perturb exchange rate against any fiat up or down by more than 0.015% per day would to be suspended or knocked out by peers (no central control involved). This way exchange rate would not change by more than 5% per year. Then let all online exchanges execute orders from same blockchain.

[knocte]

Current lower limit has been calculated at 40 bucks something /btc now due to the rising investment cost.

But if the USD exchange rate went lower, wouldn't less efficient miners just drop out, with the difficulty being lowered as needed?

How about having, not the USD as a referent, but the average of the most important currencies: EUR+USD?

[bigvern]

Wouldn't that mean that there would be somebody involved that made the decision if it needed stabilization?

[kokjo]

I don't see how it could be done with a minor change specifically on the technical side??
Elaborate.

Well, the following could be done.

Кeep all ask and bid  orders for all fiat currencies versus BTC in the blockchain. This is possible without changing the protocol. Allow peers to vote for each order, just like they vote to confirm or reject the transaction. Orders which are going to perturb exchange rate against any fiat up or down by more than 0.015% per day would to be suspended or knocked out by peers (no central control involved). This way exchange rate would not change by more than 5% per year. Then let all online exchanges execute orders from same blockchain.

and what if i traded person-2-person with a friend? whoops, your system is broken.

[yvv]

and what if i traded person-2-person with a friend? whoops, your system is broken.

You can trade person-2-persion with a friend without submitting order to the blochain. This will not affect the exchange rate.
 

[bezzeb]

Well, the following could be done.

Кeep all ask and bid  orders for all fiat currencies versus BTC in the blockchain. This is possible without changing the protocol. Allow peers to vote for each order, just like they vote to confirm or reject the transaction. Orders which are going to perturb exchange rate against any fiat up or down by more than 0.015% per day would to be suspended or knocked out by peers (no central control involved). This way exchange rate would not change by more than 5% per year. Then let all online exchanges execute orders from same blockchain.

Okay that's an interesting one yvv, thanks - made me think for a while.    But I'm already imagining how to game the system because the issue is that someone must decide 2 things externally:  What do you peg it to (i will manipulate this) and what percentages do you set (i will game this also).  Game case 1:  The rate of change between two fiat currencies changes more in one day than whatever limit the system imposes - GAME ON.  Then philosophical flaw case 2:  If i buy a bag of groceries for .2 BTC one day, how meaningless would a reference to a central fiat be?  Lastly, bad-guy abuse case 3:  I customize my local Bitcoin client to LIE on the amount of fiat i bought my BTC at.  How would other clients know I lied?  And how would anyone stop me from buying above or below the system's target rate?  I'd only need a modest sized bot-net to start pumping the market rates artificially before i make my sale and let it dump.  And silly case 4:  face to face cash BTC exchanges - how meaningless does it become then? 

So in the end, as will all other Bitcoin "improvement" ideas I've seen to date, the end result is less elegant and beautiful than what is now implemented, and the desired goals are easily circumvented leaving vast holes open for abuse.

No.  The current volatility of Bitcoin is due to its infancy and small scale.  Hell *I* see the market shift some days directly by my actions when I just sell 10 BTC.  On the smaller exchanges, selling 10BTC can swing the value by 5-10 bucks sometimes.  But I see this as a growing pain and short term opportunity for the savvy.  We traders by shuffling funds between highs and lows are facilitating liquidity and we need *a lot* more of this trading activity to nullify the chance for a quick buck via this very activity.  If there were only a few thousand people trading Dollars the same would be the case, but as a market grows, the volume will smooth things out and mature it into something stable over time which people can trust for normal commerce.  We are years off still but it's inevitable IMHO because the central banks will keep screwing people and going bust, tax payers will keep paying, and the Bitcoin protocol will just.... chug along.  We just need to sit tight while the Bitcoin ecosystem is built out and public awareness grows.  A few more banking panics will also help, but in the short term expect a bumpy ride.

For the next years until reaching maturity, BTC will always be low when the press (and public) forgets the horrible state of the global banking system, and skyrocket again when reminded.  (I'm talking to you Cyprus.)

[bezzeb]

and what if i traded person-2-person with a friend? whoops, your system is broken.

You can trade person-2-persion with a friend without submitting order to the blochain. This will not affect the exchange rate.
 

How do you trade BTC without going through the block-chain?  Did I miss the first day of school?

[SamuelSG]

and what if i traded person-2-person with a friend? whoops, your system is broken.

You can trade person-2-persion with a friend without submitting order to the blochain. This will not affect the exchange rate.
 

How do you trade BTC without going through the block-chain?  Did I miss the first day of school?

It takes trust that the other wont screw you, but physical exchange of printed (or otherwise made physical) private and public keys, to a wallet with the funds in it, can happen. Mostly for gifts, but there are physical bitcoins you can buy with tamper-evident seals over the private key.

[kokjo]

and what if i traded person-2-person with a friend? whoops, your system is broken.

You can trade person-2-persion with a friend without submitting order to the blochain. This will not affect the exchange rate.
 

How do you trade BTC without going through the block-chain?  Did I miss the first day of school?

this is not a Bitcoin thread, its a though experiment.

[kokjo]

and what if i traded person-2-person with a friend? whoops, your system is broken.

You can trade person-2-persion with a friend without submitting order to the blochain. This will not affect the exchange rate.
 

what if i write a forum post saying that i will pay 5 times the amount that the blockchain say i should, and just buy all?

[yvv]

How do you trade BTC without going through the block-chain?  Did I miss the first day of school?

You can send your BTC from your address to your friends address, and this will be recorded in blockchain. This will not affect the exchange rate if you do not place a sell order.

[bezzeb]

How do you trade BTC without going through the block-chain?  Did I miss the first day of school?

You can send your BTC from your address to your friends address, and this will be recorded in blockchain. This will not affect the exchange rate if you do not place a sell order.

Ah, that's what you meant.  Was gonna say, if it isn't in the block chain, it didn't happen!  (lol)

But still it was fun to consider - thanks for that.  Love good ideas - even if in the end I end up doubtful.  ;-)

[yvv]

what if i write a forum post saying that i will pay 5 times the amount that the blockchain say i should, and just buy all?

Sure you can do this, but this will not make your rate mainstream. To establish it, other peers will need to vote for it. You can skip a forum post and place any crazy order into blockchain. If everybody else accept it, fine, let the rate be screwed up.

[lawrence18uk]

Surely "adjusting supply" can be done by BTC holders themselves. Isn't that the whole point?

I'll put it more directly - it would be a good idea if we all played at being "central bankers", from time to time.

There will other speculative periods, with booms and busts. It's in all our interests to get as many people as possible involved in Bitcoin. When the price is rising "too fast" - there are more people wanting to buy into bitcoin than there are those willing to give up a bit of ownership. So, if "we" attempted to balance demand and supply at these critical moments we might succeed in bringing BTC price swings down below bthe critical threshold where a speculative frenzy ensues, thus ensuring a more smooth rate of appreciation - and hence BTC adoption.

In this respect "selling high, then buying low" is just a method of stabilising the currency's value.

Or, to put it another way, strict hoarders may benefit by strategic selling. Would it be possible to coordinate this? Or have some method of knowing how much is going on? No - any such information would act against the market-makers. But we could have an indication of how much market manipulation might be appropriate. Then each of us decide how much to act based on our oown holdings. Most bitcoins are, apparently, held by accounts with less than 500BTC in them, so if a lot of us acted in this way, it would have a noticable effect...

Earlier in this topic someone suggested an appreciation of less "than 5% per annum against another major currency" (basket) would be desirable. Now that seems to me to be trying to make Bitcoin like any other currency.... I would have thought that appreciation of 3 times per annum would be quite acceptable - giving appreciation of an order of magnitude (ie x10) every two years. That seems to me to be quite reasonable with a growing BTC-based economy.

[yvv]

Surely "adjusting supply" can be done by BTC holders themselves. Isn't that the whole point?

Yes, this is pretty much my point. And this can be done by voting. Whole order book for every currency in the world could be kept in the blockchain (from bitcoin wiki I understand that you don't need any changes in protocol for this). This book would be updated every 10 minutes. Every peer could explore this book and instruct his/her client on how bad disbalance between bids and asks he/she is going to tolerate, and confirm or reject orders based on this decision. If the community vote for crazy volatility, fine, let it be. If crazy conservatives win, slow down the horses. You would still need central exchanges. If all of them use same order book, approved by BTC community, they will not be able to affect exchange rates.

[yvv]

Earlier in this topic someone suggested an appreciation of less "than 5% per annum against another major currency" (basket) would be desirable. Now that seems to me to be trying to make Bitcoin like any other currency.... I would have thought that appreciation of 3 times per annum would be quite acceptable - giving appreciation of an order of magnitude (ie x10) every two years. That seems to me to be quite reasonable with a growing BTC-based economy.

If the majority approve this level of volatility, it is perfectly fine. Everybody should be able to set this limit in client settings.

[timhuge]

Folks may be interested in this related, current thread: https://bitcointalk.org/index.php?topic=178140.0

[Etlase2]

But who would supply that variable? The cost of computing? This would have to be automated and how would that be automated?

Step 1: separate currency creation from securing the network. This is a huge task and one that has been solved in the Decrits proposal.
Step 2: find a way to force miners to compete against each other in who can create coins faster by offering incentives. In addition to this, difficulty must only increase, not decrease (this can only happen if step 1 is in place). See the Decrits proposal.
Step 3: while the technical supply of coins must be unlimited, the realistic amount of coins that can be created in a certain time frame needs to be tied to network activity. To make network activity "honest", a percentage of each transaction must be taken as a fee (Decrits proposes 0.01%).
Step 4: defensive measures must be in place to protect against attacks on the difficulty via ASICs or pooled power. See deeper in the Decrits thread.
Step 4a: make mining create only a portion of new coins with the rest being given away in a lottery to reduce the "hardware tax" as I call it, so that you don't need to be a miner to make a profit using the network. This reduces the power that, for example, a region with cheap electricity would have over one with expensive electricity. It also means that few coins actually need to be paid for in hardware/electricity/time, which as we all know is a complete waste.

And billion other things that would make the network better and more efficient for fun.

<goes to work detailing specific proposal>

edit: To be clear, I'm at work and sort of booked the next couple days. But I'm planning on writing a little analysis up within the week.

I would suggest checking out Decrits before potentially reinventing the wheel. Although I mention several times in that thread that many ideas are tentative, they are not so tentative in the 6+ months of delving down and refining since.

[timhuge]

But who would supply that variable? The cost of computing? This would have to be automated and how would that be automated?

Step 1: separate currency creation from securing the network. This is a huge task and one that has been solved in the Decrits proposal.
Step 2: find a way to force miners to compete against each other in who can create coins faster by offering incentives. In addition to this, difficulty must only increase, not decrease (this can only happen if step 1 is in place). See the Decrits proposal.
Step 3: while the technical supply of coins must be unlimited, the realistic amount of coins that can be created in a certain time frame needs to be tied to network activity. To make network activity "honest", a percentage of each transaction must be taken as a fee (Decrits proposes 0.01%).
Step 4: defensive measures must be in place to protect against attacks on the difficulty via ASICs or pooled power. See deeper in the Decrits thread.
Step 4a: make mining create only a portion of new coins with the rest being given away in a lottery to reduce the "hardware tax" as I call it, so that you don't need to be a miner to make a profit using the network. This reduces the power that, for example, a region with cheap electricity would have over one with expensive electricity. It also means that few coins actually need to be paid for in hardware/electricity/time, which as we all know is a complete waste.

And billion other things that would make the network better and more efficient for fun.

<goes to work detailing specific proposal>

edit: To be clear, I'm at work and sort of booked the next couple days. But I'm planning on writing a little analysis up within the week.

I would suggest checking out Decrits before potentially reinventing the wheel. Although I mention several times in that thread that many ideas are tentative, they are not so tentative in the 6+ months of delving down and refining since.

From my initial work, I see the difficulty that you are addressing in Steps 1-3. Step 4, however, and maybe the "billion other things" seem like improvements on Bitcoin that are relatively unrelated to issue of stabilizing the value. Do you think they might be an obstacle to increasing understanding of the StableCoin idea? My thought was to keep my proposal as close to Bitcoin as possible, making only the changes necessary to stabilize the value, as a kind of proof of concept. Then other improvements to Bitcoin could be debated separately.

I am very interested in your Decrits proposal, which I just learned of yesterday, but I'm thinking of finishing outlining my own initial proposal before diving into the details. Before releasing my proposal, though, I intend to look more carefully at yours to make sure I'm not just duplicating it.