[Speaker 1]: It is February 25th, 2026. And less than two months ago, one of the most expensive experiments in the history of fintech officially came to an end. [Speaker 2]: Goldman Sachs sold the Apple Card portfolio to Chase. But here is the thing-they didn’t sell it for a profit. According to the filings and the analysis coming out of the deal, the discount was over one billion dollars. [Speaker 1]: Right. They effectively paid Chase to take these customers off their hands. Think about that for a second. Goldman Sachs-the premiere investment bank on Wall Street-wanted out of this deal so badly that they took a massive, ten-figure hit just to escape consumer lending. [Speaker 2]: For years, the marketing pitch was "Created by Apple, not a bank." It turns out there was a reason the actual banks-Chase, Amex, Citi-passed on this specific deal back in 2019. [Speaker 1]: Today, we are looking at how Goldman agreed to terms that established banks rejected, and why specific design choices-things that looked great on an iPhone screen-created a unit economics failure that cost billions. [Speaker 2]: And we’ll look at why JPMorgan Chase believes they can turn a profit where Goldman failed. [Speaker 1]: It’s Wednesday, February 25, 2026, and you’re listening to The Angle. [Speaker 2]: So let’s start with the narrative that took hold when this sale was announced in January. The headline was basically: "Goldman Sachs fails at consumer banking." [Speaker 1]: Which is true, but it misses the nuance. It wasn’t a failure of marketing. It wasn’t that people didn’t want the card. The portfolio grew massive. I mean, you saw these titanium cards everywhere. [Speaker 2]: The problem wasn't the size of the user base. It was the shape of it. In the industry, they call this a "barbell" portfolio. [Speaker 1]: Okay, visualize a barbell. Two heavy weights on opposite ends, and nothing in the middle. [Speaker 2]: Exactly. On one end, you have the "Transactors." These are the wealthy users. They have high credit scores, they put everything on the card for the points and the privacy features, and then they pay off the entire balance at the end of the month. [Speaker 1]: And banks actually lose money on those people. They cost money to service, they redeem rewards, but they pay zero interest. Usually, a bank tolerates them because they might also have a mortgage or a wealth management account. But Goldman didn’t have those other products to cross-sell. [Speaker 2]: Then you have the other end of the barbell. The subprime users. [Speaker 1]: This is where the math really broke down. By late 2023, the data showed that somewhere between 28% and 34% of Apple Card users had FICO scores under 660. [Speaker 2]: That is a massive percentage for a premium credit card. Usually, subprime lenders like Capital One or Discover have built their entire business model around managing that specific risk. But Goldman was new to this. [Speaker 1]: And the result was a net charge-off rate of 6.2%. That’s the percentage of debt that the bank just gives up on ever collecting. For context, Chase and Bank of America usually sit around 3% or lower. Goldman was running nearly double the industry average for bad debt. [Speaker 2]: So you have a portfolio where one group pays no interest because they’re rich, and the other group pays no interest because they default. There was no "middle" to generate steady profit. [Speaker 1]: How did they get there? Because this wasn't an accident. This was built into…