So, the way multi-network cards work in the US is it does not use the AID, it is all done on the merchant POS/charge acquisition software. It currently cannot be used such as for two of the main 5 or 6 networks on their own/automatically (Visa/MC/Amex/Discover fka Novus/UnionPay/Diners/the ATM/debit networks etc.) unless custom programmed by the merchant software to route as such.

Don't be mistaken, Capital One (and many other issuers including Synchrony) already do this today with their retail co-branded cards (here is a new example from Capital One). Basically, the POS software looks at the card BIN info and routes the transaction to a private network (which Capital One does have albeit relatively small for its retail partners only). This is already in action at many retailers (not just cobrands but for entire card issuers) including Wal-Mart (aside from Capital One, they also have direct routing with at least Citi and Chase bypassing a generic V/MC process), Amazon, Disney, various fuel stations, etc.

You can find my ChaseNet posts elsewhere on this forum, but ChaseNet currently runs in the Visa infrastructure (it could be separated, but would be super expensive for Chase). I believe the transaction actually still goes to Visa for auth in a sort of white label situation, but the fee schedule is entirely different and set by Chase and the merchant directly. Again, you might wonder why Visa allows this, well, you just can't say no to Chase. Theoretically, Chase takes on some of the handling and liability that Visa would otherwise have to cover, but my read is this a volume play for Chase to get more charges on their card where they can make interest from consumers to compensate for the lower cost. I am not entirely clear if Chase makes any profit on the ChaseNet offering itself especially because they have given up trying to promote Chase cards at retail. Meaning using a Chase Visa versus one from BofA, Barclays or whomever else.

And that is the situation for Capital One. Even if they move the card processing over to Discover (which I think they should), there is going to neeed to be a backup network, and I am pretty sure that Mastercard will become that backup processing network. There is no way Mastercard (or Visa) are going to turn down the chance to support transactions even as a backup partner (the ChaseNet/Visa coexistence is likely the best example). FYI, Discover has the hands-down best exchange rates overseas wherever you can use the card and beat Visa/MC every time. Yes, I understand the card acceptance issues overseas. I don't know how well this will work for the card itself. It seems the easiest way would to use the MC rails wherever Discover is not accepted natively similar to the current UP relationship, but likely still will be difficult for card acceptance technically (a lot of software updates would be needed at the merchant POS).

Finally, it is always bad news to lose an issuer, but Discover has been "for sale" for a few years, and I am sure every bank has gone through the process of determining if it wanted to make an offer for the assets. Discover has done a lot to try and make itself more attractive (getting rid of student loans, etc.), but if you get pass the merger PR (and the potential regulatory issues), here is what I see happening:

-I think the Discover brand will remain only for the merchant network (well, the old Novus really). Discover branded cards will move to the Capital One label eventually.
-While Discover has US support, Capital One has very few, I expect most of the US support to close to achieve desired cost savings unfortunately.
-I agree that the 5 percent rotating will go away. I went through the HSBC USA to Capital One process and after awhile, all of the cards that were converted eventually lost their better rebate features.

There is a lot of hedging going on based on potential legislation including alternative networks, transaction fee limits. I think this does place Capital One in a good choice particularly for offering a reasonable option in this situation, but that still seems more of a guess as I believe any economic slow-down will just delay such actions.