Re: CAP theorem
P may be largely in the electronic domain but C and A have plenty of analogous concerns in non digital money, and are subject to manual speeds.
Let's talk scaling here with respect to what's been achieved in the M0 domain to set a baseline: https://www.npci.org.in/what-we-do/upi/product-statistics
UPI had its first full year in 2017. It now has annual turnover of:
2023: $2.5 trillion from 120 billion transactions
2022: $1.4 trillion from 75 billion transactions
2021: $880 billion from 40 billion transactions
2020: $420 billion from 19 billion transactions
2019: $220 billion from !0 billion transactions
That's a near doubling YoY that's been going on several years now. Mainly driven by the ~750 million smartphone users in India.
UPI already accounts for a substantial share of RBI M0 data, and turnover will exceed the dollar GDP ($4 trillion) in perhaps 2-3 years. Of course that's normal of M0 but it's a measure of scale of digital payments.
Indian and Chinese central banks deal with mature systems operating at volume and value scale not seen anywhere else on the planet. They already have experience dealing with TD atomicity and instantaniety . Bank issued stablecoin is an incremental problem for them, and they've a prior record of building and scaling pilot designs.