The external treasury idea is okay, but a lot more complicated that in sounds to implement in practice.
I would disagree on this point:
Adding interest means creating inflation out of thin air for HBDs (if I'm wrong correct me on that, it would make my reasoning invalid). So we are going closer to the debt ceiling, much faster than in the previous bullrun, especially if this is successful and we raise interest rates even more
I don't think even higher interest rates are particularly significant here. It would take many years to increase the HBD supply by enough to matter, and that is assuming that over that period of years the increased supply wouldn't be offset by other factors (most clearly decline in demand for HBD as its supply gets too large), which is probably false. And as I mentioned in the post, a growing level of HBD is still sustainable for a very long time if not permanently, as long as the value of HIVE grows even faster. In the short term this is unreliable, but over a period of many years it is a reasonable expectation as long as Hive doesn't just fail (in which case HBD doesn't matter either).
I also think the 10% debt ratio cap should be increased. It was introduced hastily (by Steemit) without much consideration or discussion. Even in the original white paper. I believe debt ratios of 20% or 30% were mentioned as being okay, and that might even be low.
Under most conditions the market should dictate how much HBD is sustainable, since if there is too much and it is insufficiently backed by value of the underlying token, its value would drop. A cap is perhaps needed to prevent frothy "bubble" market situations from far exceeding what a more sober analysis would support, but it doesn't need to be 10%.
You are ignoring the effect proposal has. Your proposal will massively increase the amount of HBD in circulation by not allowing the price to pump.
Without it, I agree inflation is fine and wouldn't be able to ever catch up. Both the proposal and inflation together are very destructive.
We have to accept the fact that cryptos still go through bubbles. Hive collapsed by 99% last cycle. Who knows how much it will be the next time, but we have to be ready for that.
When SBD crashed form $15 to $1 in 2018 it had no negative effect on Steem. However, if we manage to sell and recycle profitably enough HBDs to keep the price at $1, an equivalent sell off will take us well below a dollar very quickly, and hive will be crashing at the same time. On top of that, there are interest rates.
We need a way to offset the negative feedback loop HBD creates.
You're assuming that a non-pumped HBD will crash as much or as easily (or at all) as one which is pumped to $10+. I believe this is false.
Yes, HBD has an amplifying effect on HIVE price changes, both positive and negative, this is the nature of leverage, and can't be "mitigated" away. You take the bad with the good. The idea is that over time, the good outweighs the bad.
In reality the leverage ratio is currently tiny (1.1x max, given the 10% cap), and even if the cap is increase the ratio will still be relatively low. Yes cryptos boom and bust, but the leverage effect of HBD is always going to be a relatively small part of this.
The benefit here is attracting net demand of capital into Hive coming from demand for yield. Again, that's a large net benefit over time, but it won't be a straight line up (and you can't make it so), it never is. There will be volatility to withstand (or not, as one chooses).
I disagree with that. The market will dictate that HBD debt is unsustainable by making both HIVE and HBD drop. Unless witnesses decide to hardfork so that HBDS can't get converted to hive anymore.
In that case they save HIVE by defaulting on debt which we are unable to repay.
Edit - The reason is that as HBD drops arbitrage starts taking place and hive drops. This is the negative feedback loop.
It is leverage. Works both up and down. Over time, predominantly up.
Calling it a "negative feedback loop" is arguably misleading, at least if you neglect to mention the "positive feedback loop" on the upside.
Yes, there is a leverage effect. That is intended. If you can find the original white paper it is discussed there.