That's a false dichotomy – there's (at least) a third option – a service that doesn't skim so much off the top – 2.5% made sense when credit cards were a rarer payment method, used only for a small fraction of transactions.
Now, nearly everyone has and uses multiple visa or mc branded credit or debit cards, yielding trillions in transactions, making billions for Visa/MC.
In short, they've grown much more profitable due to their scale and none of that has value has been returned to businesses or consumers in via rate reductions, AFAIK.
Still, money handling/cheque handling has costs as well.
Cheques are much more prone to fraud, money has some fraud cost (% of fake money, not sure how it is, but it's not that big) and costs of handling and moving the money (hence, cashback reduces this cost)
> there's (at least) a third option – a service that doesn't skim so much off the top
I don't disagree with this at all. But I don't read the comment I was responding to as having come from the thought process of
"If Visa/MC charged a 1.5% cut instead of 2.5%, the economy would be 1.5% more productive than it is now, which means that, compared to that more enlightened hypothetical world, the non-Visa/MC portion of the economy is only 97.5% what it should be."
If you think credit is related to the size of the economy (and I do), you need to ask, where did that 2.5% number come from? Saying that Visa's entire fee represents nothing but a drag on the economy is very much of a piece with the historical loathing of merchants and usurers, who, as anyone could see, did not create value.
I should also point out that, if the assumption you start with is "the drag on the economy (relative to potential) from credit card fees is equal to the amount of those fees", you should quickly notice that the directionality is wrong.
In my example example, cutting fees to 1.5% (from any level at all, interestingly) requires the economy to expand by 1.5%. That's not a coincidence -- cutting fees to 0.1% would require the economy to expand by 0.1%, except that that's completely implausible; cutting fees further should cause the economy to expand more, not less.
With that in mind, it might make sense to measure against the hypothetical where credit card companies offer their services for free, but even then there is no obvious relationship to the current level of their fees. I have to stand by my assessment that saying Visa/MC are skimming 2.5% of the economy doesn't make sense. How'd we get that number?