- Having ledger that is append only by a set of trusted parties gives you all of the benefits of a distributed ledger, as well as the ability to do things like clawbacks, comply with regulation, etc.
- The future I see is multiple blockchains operated by networks of payment processors. For instance, Visa, MasterCard, etc. can all work together to maintain one blockchain. Other countries / regions can have their own. Interop can be done with bridges (like in defi now).
- Throughput can be drastically increased between banks. Transactions become confirmed instantly. No more waiting several days for ACH. You can confirm a good balance from the sender, and credit the receiver instantly.
- "Just use a database" the hackernews commenter shouted out. Using a database instead of a blockchain limits what you can do in terms of 3rd party interop, and hamstrings you for upgradability in the future. Facebook's MySQL upgrade is taking YEARS, and they're a SINGLE TECH COMPANY. If you want to move faster, and maintain interop blockchain is the way to go.
- This opens the door to multi-asset blockchains between financial institutions. Instruments like loans, equities, etc. can be tokenized, FRACTIONALIZED and traded b/t institutions much more quickly than they can today, because they'll be a part of a permissioned verifiable system. Write a loan today, fractionalize and sell it instantly.
There's so much more that's possible here too. Technologies like HyperLedger are really moving the ball forward. I think a lot of the criticisms of this sort of thing are due to shortsightedness of what is possible.