What I have in mind is something a little bit like the zerocoin proposal, in the respect that it’s an alternate coin which can be translated to Bitcoin and back as part of how the system works.
But instead of being fixed with respect to the bitcoin it would increase or decrease in relative value in response to market signals. There would be an exchange rate determined by dividing the number of spendable units of alternate currency against the number of bitcoins locked behind it.
So what you need is a series of rules for controlling that number that tend towards economic stability. Which is itself kind of tricky. I had thought locking the coins by a time-lock could do it, but I haven’t thought of a way to base an exchange rate on that. Empirical work with simulator or a smart math person with a whiteboard might turn up something.
The following uses incentives to accomplish something similar-in-principle:
Alice creates 100 dollarcoins at a cost of 1 bitcoin. She transfers 1 to Bob, he can turn it into 0.01 bitcoin.
The dollarcoin has a specification that includes things like interest payments. If Bob waits 30 days after receiving a dollarcoin he gets 1% of its nominal bitcoin value in interest. He gets another 1% payment 30 days later, and so forth. But if he spends it right away or converts it to bitcoin there are no payments.
Claire decides to use the dollarcoin specification to mint some dollarcoins. She has to pay a 1% minting fee, so 1.01 bitcoins to make 100 dollarcoins. The backing rate becomes 200 dollarcoins to 2.01 bitcoins.
Thus the more dollarcoins get minted the more valuable/highly backed they are, initially. If people spend them quickly, or convert them to bitcoin quickly without collecting interest, they continue to grow in bitcoin value. On the other hand if people hold onto them over time, they become less valuable as the savers get paid for keeping them.
Criticisms: It’s complicated, and the numbers for interest rates and minting fees are pulled out of thin air. Presumably there’s some range that works well, and could be tested by implementing multiple currency specifications side by side.
What I have in mind is something a little bit like the zerocoin proposal, in the respect that it’s an alternate coin which can be translated to Bitcoin and back as part of how the system works.
But instead of being fixed with respect to the bitcoin it would increase or decrease in relative value in response to market signals. There would be an exchange rate determined by dividing the number of spendable units of alternate currency against the number of bitcoins locked behind it.
So what you need is a series of rules for controlling that number that tend towards economic stability. Which is itself kind of tricky. I had thought locking the coins by a time-lock could do it, but I haven’t thought of a way to base an exchange rate on that. Empirical work with simulator or a smart math person with a whiteboard might turn up something.
The following uses incentives to accomplish something similar-in-principle:
Alice creates 100 dollarcoins at a cost of 1 bitcoin. She transfers 1 to Bob, he can turn it into 0.01 bitcoin.
The dollarcoin has a specification that includes things like interest payments. If Bob waits 30 days after receiving a dollarcoin he gets 1% of its nominal bitcoin value in interest. He gets another 1% payment 30 days later, and so forth. But if he spends it right away or converts it to bitcoin there are no payments.
Claire decides to use the dollarcoin specification to mint some dollarcoins. She has to pay a 1% minting fee, so 1.01 bitcoins to make 100 dollarcoins. The backing rate becomes 200 dollarcoins to 2.01 bitcoins.
Thus the more dollarcoins get minted the more valuable/highly backed they are, initially. If people spend them quickly, or convert them to bitcoin quickly without collecting interest, they continue to grow in bitcoin value. On the other hand if people hold onto them over time, they become less valuable as the savers get paid for keeping them.
Criticisms: It’s complicated, and the numbers for interest rates and minting fees are pulled out of thin air. Presumably there’s some range that works well, and could be tested by implementing multiple currency specifications side by side.