Idea for stable-value cryptocurrency without centralized authority!

[TheKing01]

Some currencies have tried to be stable by having a centralized authority which backs up the coin for something of worth. The problem is, sometimes these centralized authority spontaneously evaporate (https://daology.org/proposals/dcda223aa190358023cc066208f820614df3f310).

In https://bitcointalk.org/index.php?topic=1466607.msg14800985, I described an idea for a pegged cryptocurrency, based off another coin for arbitrage. The problem was that getting information from the outside world is really hard. I realized that we do have information from the outside world: the difficulty.

In particular, the difficulty reflects the value of the coin being mined.

Therefore, I propose the following:

There will be two coins, StableCoin (or just SC) and VolatileCoin (or just VC). StableCoin will have a stable price. When there is a price pressure on StableCoin, VolatileCoin's price changes instead.

How is this accomplished? Miners will mine VC, getting some reward each block. They also get transaction fees. The smallest unit of SC is the hash (similar to the satoshi in bitcoin). The average number of hashes miners have to do mine a block can be determined from difficulty. An exchange rate (or ER) is determined, by setting hashes = the miner reward (it's okay if some or all transaction fees are in SC, you just need a little bit of algebra.)

What is the significance of ER?  StableCoin can be converted on the blockchain to VolatileCoin according to the ER, and vice versa, using a special transaction.This isn't simply a trade; if you convert SC to VC, SC is being destroyed, and VC is being created.

What this means is that the actual exchange rate on markets between SC and VC must be ER, or arbitrage will occur to correct it.

A hash will always be equal in value to the amount of electricity required to do a hash (which we presume to be stable). Since a hash is just a unit of SC, that means SC is also stable. The only way for this to fail is if the value of VC falls to *exactly* zero, since this would cause a division by 0 error.

Some notes: You must make sure that no coin is merge mined with VC, or miner reward (in VC) won't equal the number of hashes. One thing you *can* do though is make it so that the hardware used to mine VC can be used to mine other coins (such as BitCoin), but not at the same time. This makes is easier for miners to respond to price changes in VC.


So, what do you guys think? Would this work? If not, in what manner exactly would it fail (keep arbitrage in mind). I haven't taken any Econ courses, so I'm wondering what someone who has will think of this idea (I have some ideas about how this coin might fail, but I'm sort of a fish out of water without basic Econ experience.)

Also, if anyone wants to steal this idea, go ahead. If no one does, and no one finds any obvious flaws in this design, I may look into actually starting this coin. Having stability in your currencies value is a huge win, and is necessary both for long term savings and day to day use. Although you can achieve stability by having a big economy, the above shows how you can "cheat" the system and get stability with a relatively small economy potentially. (If you do steal the idea, PM me. I would love to see what actually happens.)

[TheKing01]

Some further notes:

One problem is that the price of mining may change (most likely fall). This can occur either due to more efficient mining rigs, or lower electricity costs.

First is the problem of more efficient mining rigs. One way to solve this is to make sure they are already as efficient as possible, choosing a hashing algorithm that has already been fully optimized.

A more sophisticated approach is to use multiple, separate hash functions, such that they are unlikely to optimized at the same time. We could imagine each having their own stable coin (on the same block chain), and when a hash function is about to be or is being optimized, those coins can be destroyed in favor of another hash function until the optimization is done. Since this isn't trading, but conversion, this shouldn't cause any economic problems. (When mining, you are always rewarded in TradeCoin, regardless of the hash algorithm you are using). This would also be interesting in that different blocks would have different proofs of work.

An even more sophisticated modification of the above is for the ER between VolatileCoin and StableCoin to be based on multiple hash functions. When the difficulty of one of the hash function rises rapidly relative to the others, you know that means that it is being optimized, and the network can compensate. Once it stops rising, it can be re-pegged to StableCoin. This has the advantage of having only on StableCoin (a good thing for cold storage users) but the disadvantage of requiring a more complicated algorithm (which may not work as well as a market).

Concerning the multiple StableCoin proposal, you could introduce a third key called the "Conversion Key" that can convert but not spend coins associated with an address. This way for cold storage you only need to keep the private key really secure, since the Conversion Key is a less profitable target.

The harder problem is electricity costs. Although electricity has been better at keeping value than fiat currency, it is still decreasing. One thing one could do is put is hard code a factor to adjust the ER based on how we predict the value of electricity to change, but that is probably hard to predict (it could rise in value once fuel starts running out).

We could try repeating the above perhaps. In addition to an electricity based POW, you could have storage based POW. You can have people sell their disk space, ala Storj and Sia, for TradeCoin. This can help predict the correct ER.

Concerning short term stability ("wiggling in the price") although shorter times between blocks, advanced difficulty adjustment algorithms (maybe something like PID) and lower difficulty uncle blocks can help make sure the ER is up to date, I don't think that is necessary if long term stability is achieved. If the price falls a little bit, no one would sell, since they would just wait for the price to go back up. If the price falls a bit, no one would buy, since they would just wait for the price to go back down. For most assets, like other cryptocurrencies, you don't know if the price will go back up or down, but if long term stability is achieved, you do know. What would happen is that if the ER is adjusted fast enough, VolatileCoin would change in price, not StableCoin. (Again, this paragraph is dependent on the assumption that StableCoin's price is in fact stable (or at least predictable).)

[Nascor]

there are already other alternatives out there, like "baskets of currencies", they work like derivatives

their value derives therefore from the underlying asset, which could be fiat money for example which is already stable

[krigger]

Some people are saying Bitcoin is becoming centralized because of SegWait, although I do not know enough about the technology to really comment on much detail about that.