I was trying to keep it short and simple by skipping a step but yes, the SSA follows a formula to raise the cap. But anything the executive does must be authorized by Congress, including the current formula which was set in a reauthorization bill back in the 80s (I think, maybe the 70s, apologies, but I'm not able to look it up right now). So far, every time a budget is passed and every few years when the SSA needs to be reauthorized, they've left them alone. Despite the occasional bill messing with the SSA getting introduced, they never get out of committee.
As far as the CBO goes I don't recall ever reading about cap increases in their report summaries on the trust fund. Although I have read their reports on the effect of various proposed changes to the way the cap is calculated. I'll have to do some more looking when I have the time, but I was definitely under the impression cap increases were in a category the CBO didn't anticipate future changes to when evaluating the health of the trust fund. I thought normally the COLAs would also fall into this category but that is overridden by them being mandatory spending, as opposed to discretionary, so they have to be taken into account. I'm certainly no expert and wouldn't be surprised to find out I missed something.
Have you not been paying attention to what the Fed has been doing? Pardon my language but shoving cash up my asshole earns more than 1% these days.
In 2022, before most of the rate hikes, the trust fund earned $66.4 billion. This year's high, and hopefully very temporary, interest rates aside, it'll usually be around 2.5-3%.
I'm not sure what you think loss leading means or why you're using it here, but governments storing reserve money earmarked for a specific purpose in their own bonds isn't unusual or a bad thing. Should they stuff it under a mattress earning 0%? Should they risk it in the markets? Unsecured domestic bonds? Foreign bonds?