>That model is not analogous to the credit card situation, in multiple ways. Among other things, it's framing this as a "tax" (which isn't inherently the right model), and presupposing that the origin of the "tax" is the credit card interchange, and mapping the "rewards" programs to a discount on the "tax" but not mapping anything else (e.g. free checking or the availability of credit instruments that wouldn't otherwise be available) to that, with a lot of assumptions about which parts of the overall system to include and map, and which parts to leave out. The net result seems like a cherry-picked conclusion to fit an agenda. If you decide in advance what you want the model to show, you can make a model to show it, but that doesn't mean that model is an accurate representation of the system.

That's a lot of words, but I don't see how it refutes the core point which is that "rich" cardholders pay 1% (or whatever) less on their spend than someone paying with debit or cash. All you did is handwave a bit about how interchange fees aren't really like a tax, and how the logic is "cherry-picked".

>Card companies/issuers charge interchange so that the credit card company makes money; they don't do it with the primary goal of funding rewards programs, or free checking, or the other things they do for marketing purposes. That would be like saying "the primary reason this company charges for their product is to spend money on marketing programs". Credit card companies didn't pick their interchange rates on the basis of funding reward programs, specifically; they set their rates to make money for themselves.

Again, this is a lot of words but I don't see how this refutes the claim that rich cardholders get 1% back but poorer people paying with debit/cash do not. Moreover, if you're sufficiently cynical, you can claim that the government levies taxes so they "make money", not "with the primary goal" of funding schools and roads.

>Also, to the best of my knowledge, current law no longer allows credit card companies to prohibit merchants from charging a premium for using credit cards, or for using specific credit cards. (Credit card companies used to do this, which effectively made them a cartel engaging in price-fixing.) e.g. there is nothing preventing merchants from charging less to people with cards that cost less to accept, such as debit cards or less "premium" credit cards. In theory, doing so might create competition for cards with lower interchange, or incentives for people to stop using rewards cards. In practice, however, merchants don't do this. Given that, you could just as easily portray this as a model where merchants are choosing to value the custom of higher-income people (e.g. because they spend more) over the custom of lower-income people. I don't think that's an accurate model either, though.

The fact that merchants are freely choosing to give rich cardholders subsidies doesn't diminish the fact that rich cardholders are being subsidized. It might be better than some imaginary system where they're forced to subsidize rich cardholders, but detractors of cashback/rewards programs oppose such programs existing at all.

>Credit cards make money from both low-income and high-income people alike, and make more money from high-income people, and neither one is subsidizing the other.

You're committing the same mistake that you allege me doing above (ie. "If you decide in advance what you want the model to show, you can make a model to show it, but that doesn't mean that model is an accurate representation of the system."). In particular, you're restricting yourself to only analyzing the revenue/expenses from the card issuer's perspective, and not analyzing how much the customer ends up paying. It's possible simultaneously for a card issuer to be making money off of rich people, and for poor people to be screwed over by the interchange fee system. An overly simple model that demonstrates this would be a population divided into "rich" and "poor", where "rich" people use credit cards with 1% cashback and 2% interchange, and "poor" people use credit cards with 0% cashback and 0.1% interchange. In this model, from the perspective of the bank, they're clearly making more money off "rich" people in both absolute and relative terms (2% - 1% cashback = 1% profit, compared to 0.1% interchange for "poor" people). However the rich would still be paying a lower effective price for whatever they're buying at the stores.

Of course, this analysis leaves out a bunch of details, but neither Patrick's thread nor your comment tries to refute why the model above is wrong, why we shouldn't use "effective price" (ie. price paid - cashback) as the thing to analyze, or we why we should focus on some other metric (eg. card issuer profit) instead. All he did was point out some other metric and say "but these metrics say they're making money off rich people as well, so you're wrong!", without trying to refute the original claim. It's like arguing with a "replace income tax with tariffs" proponent, and having him respond to your claim that tariffs are regressive with "yeah but rich people still pay more in absolute terms so it's not regressive!".