I think that over time the amount of SBI that the voting accounts will be able to provide is likely to increase over time
This is my expectation too. Hive SBI vote values limited by the maximum deliverable on each account. We target 50-70% VP across all accounts and are pretty good about keeping in that range, as long as the nodes that we connect to are cooperative. I would like to run my own node at some point, but we don't have the budget for that yet.
We keep a cushion in both HBD and HIVE so that we can capture value from the conversion features that help HBD to maintain its peg, but generally power up 500 to 1000 HP per week. In addition there is ongoing curation income that keeps all the accounts steadily growing. Over time that means that the maximum deliverable upvote will gradually creep up (assuming steady HIVE price). If HIVE goes up again, even better! (Although rising HIVE prices tend to bring back inactive members, so the growth in our upvote values can lag HIVE price growth a little, since returning members put additional pressure on our VP.)
It's all behind the scenes complicated (though it's actually a few simple rules that I follow that result in behavior that looks a lot more complicated than it actually is).
Thanks for the reply
I have question for you, if I send SBI to @null, what happens there? Is that a case where their balance will just increase forever? But if that's the case why isn't there balance higher? I just looked and they have around 300 units.
Its really interesting how easy it is to create complex systems with only a few changing variables. I really don't think a lot of people appreciate that.
We run a script periodically that caps pending balance at one year of accumulation for accounts that have been inactive (no posts or comments) for longer than that.
Effectively you stop accruing after one year of inactivity, but it's much less resource intensive to scrape of the excess periodically than to check every account for active age every accrual cycle.