That's correct, but I meant to ask this: How is your proposed grant better than the following example, where a fee is paid to bring a custodian's own funds to the peg? (9000 NBT is the mined amount in this example)
1) Grant a custodian 9000 + 25% NBT, which they keep
2) One week later (to simulate the mining period), the custodian puts BTC worth 9000 NBT of buy support on market
3) As that BTC sells, the custodian moves the NBT to the sell side of the market
4) As that NBT is sold, the custodian prepares dividend payments
Is this model, in effect, the same as your proposal, but without the mining variability risk/reward? Note that this leaves shareholder funds in custodian hands for a possibly lengthy amount of time, because the wall(s) may take a long time to fully sell.
Now, compare that model to the following:
1) Grant a custodian 25% of 9000 NBT in fees (but not the base 9000 NBT amount)
2) Custodian brings 9000 NBT-worth of BTC of their own funds to the buy side immediately
3) Custodian supports the peg under the same plan as before, but this time, keeps the funds at the end instead of distributing dividends. Buy wall -> Sell wall -> Withdraw support.
I could be wrong, but I think this simpler model is the same peg support mechanism, except with immediate support instead of one-week-delayed support, and without the extra 9000 NBT granted and paid as dividends. In a strange twist, I think this could be made equivalent with the following final step:
4) At the conclusion of the support lifetime, issue a second grant of 9000 NBT to the custodian, which they keep. They immediately pay out 9000 NBT worth of dividends out of their own funds.
Is this also equivalent to the draft proposal under discussion (but without the extended period of dividend-bound funds under custodian control)? Please correct anything I've overlooked.