I would say the fraud protection from T+2 is very minimal and there are better ways to implement reversible transactions. Nearly all of the large banks have the

I would say the fraud protection from T+2 is very minimal and there are better ways to implement reversible transactions. Nearly all of the large banks have their own instant payment methods to send cash to friends. If you receive alerts on withdrawals & transfers those happen nearly instantly so if you are going to notice a suspicious transaction it will either be the same day or at the end of the month when people log in to their bank account. Plus, you have to hope that the bank/etc. is going to actually use the T+2 delay to your advantage. A fairly easy way to add fraud protection would be to have T-0 or instant settlement only for previously used accounts/routing numbers. A lot of services already implement this on their end such that new bank accounts take a few days or a week to transfer but future transfers appear instant or within a day. Other options include allowing the first $X-thousand to be instant and the rest settles in T+2. This is pretty common if you get automatic deposits over a certain amount or use a phone check scanning feature. I guess my point is that a lot of edge services (services that people interact with) already cover the T+2 delay in their own manner but that leads to confusion when some middle service provider decides to change the rules. Making the system itself adopt those strategies would just distribute the risk not increase it.