Author Topic: Potential Market Manipulations on BitAssets  (Read 3605 times)

0 Members and 1 Guest are viewing this topic.

Offline toast

  • Hero Member
  • *****
  • Posts: 4001
    • View Profile
  • BitShares: nikolai
Prior discussions, covers #2 and #3 especially well I think:   https://bitsharestalk.org/index.php?topic=3130.0

Also I think #1 is not an attack but is expected behavior. You should get rewarded for trading towards the true price and you should get punished by trying to sell undervalued.
Quote
We can see when shorts are about to be forced to cover in the market due to expiries. Its possible in this situation to buy bitAsset just prior to the cover at the lowest ask, then set a sell order a distance into the sell queue where it is guaranteed to be picked up by the volume of the short-cover. As the sell price will be higher, a risk-free profit results, at the expense of the short.
Do not use this post as information for making any important decisions. The only agreements I ever make are informal and non-binding. Take the same precautions as when dealing with a compromised account, scammer, sockpuppet, etc.

Offline starspirit

  • Hero Member
  • *****
  • Posts: 948
  • Financial markets pro over 20 years
    • View Profile
  • BitShares: starspirit
In thinking about arbitrage risks, I've come across a number of situations where bitAssets could potentially be manipulated. I say potentially, because there may be reasons why such attempts either don't make sense economically, or are prevented by a rule I've overlooked, or are just practically infeasible. Nevertheless, they might be worth discussing to ensure the resilience of bitAssets. Even though I was a bit reluctant to map out the crimes, Agent86 suggested to me that it would be better for an open discussion. This is not meant to be a comprehensive list, and others may present similar situations to be alert to.

(1) Front-running of expired shorts

We can see when shorts are about to be forced to cover in the market due to expiries. Its possible in this situation to buy bitAsset just prior to the cover at the lowest ask, then set a sell order a distance into the sell queue where it is guaranteed to be picked up by the volume of the short-cover. As the sell price will be higher, a risk-free profit results, at the expense of the short.

Some possible ways to mitigate this manipulation - (i) if enough market participants try to benefit from this, they will bunch their sells very close together, lowering the cost to the short (ii) it might be possible to randomise the timing of expired shorts over some window to make it riskier for front-runners, though this makes the rolls harder for shorts to manage

(2) Forced short-squeeze

If the BTS market is less liquid than the bitAsset market (which at some point in future could conceivably be the case with bitUSD), then a large enough player could force the BTS price down substantially, and conceivably trigger an amount of margin calls on shorts that overwhelms the ability of the current sell/short queue to absorb it. If this player sat in the market with an incredibly high sell bitUSD order (e.g. on $1 of bitUSD), knowing it must be hit, they would effectively win nearly all of the BTS of these shorts held in the pool for next to nothing. This in turn could leave remaining bitUSD holders significantly under-collateralised.

Some possible ways to mitigate this manipulation - (i) don't allow margin calls on shorts to force-buy at more than a limited premium above the peg, as any higher than this actually reduces the collateral protection for remaining bitUSD holders, defying the point of margin calls in the first place

(3) Bank run

If the bitAsset became suddenly under-collateralised (e.g. if BTS collapses substantially in a short period), then margin calls are forced on nearly all remaining shorts. Anybody sitting in the sell queue at that time gets paid full value on their bitAsset (or potentially much better), which means that the collateralisation ratio on the pool falls further for the remaining bitAsset holders. If it does not play out all at once, but as a succession of BTS falls forcing waves of short-covering, the process continues until the BTS pool is all consumed leaving the last bag-holders with nothing. The analogy I draw is a bank-run. During such an event, there is also much greater vulnerability to players trying to pull off the short-squeeze above, because the required push on the BTS price is less to trigger the margin calls.

Some possible ways to mitigate this risk - (i) don't allow margin calls on shorts to force-buy at a price beyond which the collateral pool is unfairly compromised for remaining bitAsset holders - this would be the fair settlement price if the market were settled up (ii) close the market in the event of an under-collateralisation event - settle/reset the market, or consider another mechanism for replenishing the pool (?)

(4) Collusion by free BTS holders

If bitAssets grew dramatically, but BTS market cap did not grow as fast, its possible the BTS inside the collateral pool increases to a large percentage of all BTS on issue. This BTS cannot be traded on exchanges. In this case the owners of the remaining free-float BTS have a greater chance of colluding. For example, they could buy the bitAsset, withhold their BTS from being used to create more bitAsset, force the bitAsset to a high premium over the peg and sell it. The BTS market on the exchanges would also be less liquid, so they might also attempt the short-squeeze above at an opportune time.