I agree that out of this, a stronger crypto market will be born, but I'm not sure I understand your rationale.
An exchange shouldn't lose anything just because a person decides to take their coins into cold storage. Perhaps you mean the defi projects where people are earning yield? If this, then for sure that is the case. At the same time though, if the only reason to hold crypto was for the yield, then I can see more selling as the only use case for these people was the yield.
Bitcoin of course doesn't really offer any yield, its just an asset, but other coins that you can either get yield on, or staking benefits, then this of course qualifies.
But I fully agree that once the flush out happens, the crypto will be in stronger hands.
Yea, mainly the crypto firms offering yield and doing rehypothetication and the cryptos are borrowed by traders to short on exchanges via futures
It's actually a good business model and gives value for cryptos to be used as collateral to buy a house or a car. I used BlockFi to get a loan before and my experience was positive
But in times of stress such as the one we're currently in, duration mismatch (thanks MrMuppet) is catastrophic
When the cryptos depositors expect to withdraw their assets 24/7 to their local wallets in a few minutes, and the borrowers can simply hold on to the cryptos as long as they want per their terms, there's just no possible time agreement between the 2 parties and the centralized platform is forced to become insolvent
Add to that the exposure to overleveraged traders like 3AC...
Anyway, maybe a future version of BlockFi or Celsius would be ones where depositors' withdrawals of crypto assets will take 10 business days and the borrowers are compelled to return the cryptos assets within 5 business days upon demand. This would be similar in the past to requesting stock certificates to prevent the stock from being shorted
But in general, more cryptos in local wallets mean less circulating supplies for sale in exchanges and help the price to be stable or have an up bias
