bitshares / bsips

BitShares Improvement Proposals and Protocols. These technical documents describe the process of updating and improving the BitShares blockchain and technical ecosystem.
https://bitshares.github.io
63 stars 86 forks source link

BSIP41: Reduce MSSR of bitCNY from 1.1 to 1.05 #98

Closed abitmore closed 5 years ago

abitmore commented 6 years ago

PR for #97.

xeroc commented 6 years ago

I would like the summary to include more of the concerns raised in the forums. Specifically:

MSSR has one specific purpose: to incentivize shorters to avoid being margin called (or rather to punish them if they are). Margin calls are bad for the health of our smartcoins and should be avoided. Shorters are expected to actively manage their positions so they are not margin called.

It is clear, IMO, that bitCNY shorters are doing a very bad job here. As can be seen here (https://wallet.bitshares.org/#/asset/CNY ), most shorters make use of the recently introduced target_collateral_ratio to keep their collateral at the bare minimum. Obviously, they prefer being margin called. At the same time, they are rewarded from the settlement offset of 5% everytime someone settles a long position!

The logical conclusion is that the MSSR must be increased so its original purpose is fulfilled.

I agree that the margin call as such is neither good nor bad. It is, however, an indication of a bad (read: dangerous) situation, i. e. short positions without sufficient collateral, which increases the likelyhood of a black swan.

The health of the system depends on shorters providing sufficient collateral. In bitCNY, shorters don't do that. They prefer being margin called. They even deliberately try to keep the collateral as low as possible, perhaps because only the least collateralized positions can profit from the 5% settlement offset.

IMO this is an indication of a very unhealthy situation. The settlement offset is an incentive for shorters to keep their collateral as low as possible, and the current MSSR isn't sufficient incentive for providing enough collateral. Your suggestion increases the likelyhood of a black swan, as you correctly pointed out above.

Additionally, I would like one thing to be added and that is a time limitation for this expermient after which another vote will be required to keep this BSIP active. The purpose is to put the community at ease with this experiment and not have an open-ended experiment like the 5% offset we have on bitCNY.