If arbitration is the reason these companies exists, it seems like bad deal.
Maybe they sell an illusion?
Have you considered that the possibility of disputes and the existence of a framework for resolving them is part of the reason for that?
Some of it is spent by the issuer on fraud expenses that they are assigned liability for, but given the same reasoning as above, if that was more than 0.05%, there wouldn't be any profitable debit issuers left in the US (and similarly for the EU, although the regulator is changing the fraud calculus there significantly by enforcing strong cardholder authentication, so it's not an apples to apples comparison).
In other words, if there was political will to get rid of credit card points in the US, we could have all of this for much cheaper. (We might need to look into scheme fees too while we're at it, e.g. by finding a market solution that creates actual competition there.)
The world is full of services, valuable, difficult to provide services, that a given person will never consume. Doesn't mean no one else does.
It's not a value proposition from gmail that you can't access my inbox just from knowing my address.
I can't believe you can be a $400B company and the value comes from stopping fraud that you enabled by your own product design.
Imagine if you were defrauded for a few hundred bucks by someone claiming to be a big celebrity. You probably wouldn't even be able to get into contact with the actual person, let alone have them jump through a bunch of hoops, spend time on hold with your staff or go to the post office to send you copies of documents. They wouldn't bother to respond to you, they have no obligation to, whatever happened had nothing to do with them.
And now the bank pushed me to use Zelle - which has zero protection.
Also channeling patio11 [1]: This is actually a pretty interesting ongoing experiment as to whether banks can opt out of Regulation E.