When I calculate what Jim calls "the implied interest rate", it is generally lower for a BRK call than the similar SPY call.
That's why BRK over SPY.

This is true, but it's actually a reason that the income from writing calls against Brk isn't very high, making SPY the better choice on that one test. To maximize rate of return from premiums, you'd want something with rich premiums, normally something with a volatile stock price.

But as both of us mentioned, that's not the overriding criterion. Most of all, you want an idea of which way the price is most likely to move. Berkshire's price is not that hard to predict: extrapolate the likely growth trajectory of observable value, slap on a typical multiple for the end date, and the error bars are usefully small.

Conversely, despite many years of analysis in the area, I don't have a clue what the S&P might be at a year from now. The mean reversion is just way too weak, and the observed extremes just too far apart. It's like an ant wandering around somewhere in the middle of Madagascar. You know his range is ultimately bounded, sure, but you have no idea at all whether his next move will be a mile north or a mile south.

Jim