Here's an interesting (but dated) one... from the amazing Substack "Net Interest."

I had occasion today to look through point and reward economics, and next to central banking or sex trafficking or whatever it must be one of the most despicable industries out there. It's well-known that loyalty cards/rewards schemes are fiat money-printing scams that, at best, saddles the user with a below-mininum wage job (#869415).

...but it's good for the companies running these schemes. (Yea, no shit: the money printing/issuing was always great, and under shitty monetary regime there's an incentive for everyone to become a bank or a bank-like entity #743123, #1329539)

It's a typical example where businesses do things against the interest of their customer (thus, it's dishonest) while looking like they're providing you with extra benefits. Somebody should call Lizzy (Warren!). (Also, it's one area where the European Union, in its infinite overlording wisdom, made a limiting regulation that was actually good for us Europooreans! By capping interchange fees, they basically outlawed rewards programs and saved/spared 500 million Europeans the time/necessary effort to play these silly money games.)

In the age of card networks and (FX) spreads (#1586158) and costs to merchants, the COST and BENEFITS of these (Ponzi?) schemes become entirely intransparent where everyone, like in democracy, think they might get one over on someone else.

At best, it's a selection-biased gamble where sophisticated consumers take advantage by optimizing the terms/conditions/value of the whole scheme vs naive consumers who mistakenly believe they can/will win, while the house (as every properly set-up gamble) wins. The naive consumers will inevitably make some sort of error in their financial behavior (overdrafts, not enough spending, not maximizing rewards, too small fish such that annual fee eats up all other benefits).

The most well-known examples here are

• bank deposits: people hold too much cash at low-yielding bank accounts vs what they could/optimally should keep in higher-yielding savings accounts. The difference, while maaaybe sometimes convenient for the customer, end up in aggregate securing a massive windfall (=cheap financing) for the banks
• life insurance products: life insurance is usually priced very affordably and attractively for the buyer, which the insurance company isn't doing out of ignorance or stupidity but because they know from experience that some portion of the buyers (accidents, life circumstances changing, or just neglect) will let their policies lapse after having paid for a few years, meaning free money for the company, or worse, sell it back to the company at a great discount (=again, free money)

Here's the conclusion from a Fed paper on the subject from a few years ago:

regardless of income, sophisticated individuals profit from reward credit cards at the expense of naïve consumers. To probe the underlying mechanisms, we exploit bank-initiated account limit increases at the card level and show that reward cards induce more spending, leaving naïve consumers with higher unpaid balances [...]
Naïve consumers might underestimate these payments and incur costs from usage. Sophisticated consumers, in contrast, might rake in the benefits while avoiding the payments and thus profit from usage

Here's the graph they use to display the idea (sorted by FICO scores, but the same point stands)

tl;dr, the area between the curves is larger on the negative side than the positive side, thus a) redistributing from naive to sophisticated, but more importantly b) makes the entire scam profitable to the company to run: What it has to pay the sophisticated winners is less than what they manage to extract from the idiot losers.

The entire edifice is a combination of beggar-thy-neighbor and race to the bottom, with misperception and money illusion becoming key features.

From our pal Matt Levine in a recent Money Stuff piece

banks make a lot of their money from depositor inattention: When interest rates go up, bank customers mostly don’t move their money, because it would be a pain, so banks can pay below-market interest rates on their deposits.
You could have similar worries about life insurance, or annuities, or mortgages: Lots of financial products are built (and priced) based on assumptions about customer inattention.

The strategy is that you intentionally misprice (underprice) your service, banking on the fact that, in aggregate, enough people who opt in for this juicy, nice-looking offers will make enough errors that the few sophisticated customers who take perfect, optimized advantage of the mispricing won't cost the company.

ANYWAY, Marc Rubenstein walks through the logic and economics of loyalty programs and points(!) to some of the intricate cross-subsidized and legal/economic oddities with these programs:

A key question in the whole points dynamic is how much partners pay for miles. Avios Group deferred revenue at a rate of 0.88 pence per point on the stock it issued in 2023, its highest rate in many years. The consumer value of points varies widely according to how they are used, but The Points Guy puts the value of Avios at around 1.2 pence per point, so the benefit to the consumer is clear. 
Because the airline can adjust the redemption rate to encourage the use of rewards in ways that reduce their marginal cost, it is able to offer rewards to card issuers at a discount.2 In this way, reward programs can generate profits for both issuers and airlines while generating loyalty to both, a significant positive sum outcome.

“By minting points, airline companies are able to lock in substantial cash flows.”

Online points-monitoring sites posting point “valuation” tables on a regular basis because of course the offers change and of course the issuer of this fake, unusable money manipulate it to their advantage.
The only problem is that competition has been driving up the rate at which card companies pay rewards. Over the last 12 months, American Express has paid out 1.07% of purchase volume in rewards, up from 0.80% ten years ago. Other card issuers have seen similar reward inflation. Capital One paid out 1.29% of purchase volume in 2022, up from 0.80% in 2013. And Discover paid out 1.43%, up from 0.94%. 
American Express was initially exempt from the EU rules, so my British Airways co-branded card remained unaffected. In 2018, though, the courts brought Amex into line. Fortunately, American Express agreed to eat the cost. My BA Premium Plus card continues to pay 1.5 Avios points for every £1 spent, equivalent to around 1.8% on the basis of The Points Guy’s 1.2 pence per Avios valuation. I guess I do pay £250 per year for the privilege, and Amex likely bought the points at a steep discount (in particular the batch they acquired in 2020 to help the airline out of a liquidity squeeze).

JUST that gives us a few intransparent economic flows:

• First: The Amex bailout of BA during ‘rona, purchasing made-up points at cheap discount… even if Amex nominally loses ~1.3% per dollar of EU transactions, if what they hand over to the customer in rewards (=made-up BA money) was acquired at well below par, there’s still a profit in it for the card company.
• Second, to recoup the yearly card fee of 250 quid, a user needs to transact just below £14,000 a year, basically most of the discretionary card spending for a normal (Europoorean) person… and that’s not giving you pounds back, but made-up BA money you must do some additional trixing (in their webshops, for air fares, for lounge access or extra luggage etc) to benefit from.

“there are clear subsidies at play. Someone, somewhere is subsidizing my points habit – no longer the merchant, it is now other BA or Amex customers. Indeed, such subsidies are the most controversial aspect to the model.”

I don't quite understand what this percentage is referring to (a larger share of Amex revenues are going back to users than before...?) but it seems to suggest that there is some sort of competitive pressure to close this rent-seeking gap