I think that the huge equity portfolio will morph into an index fund, or mirror the SPX in some way.

I think that it's quite possible for them to do meaningfully better than the S&P 500 even with very very large portfolios--an long run advantage of 1%/year might be possible. Not by picking winners and concentrated positions in them--that's really hard unless your name is Buffett. But by merely skipping the ones most likely to be losers and playing the statistics game a bit. It's amazing how much better a portfolio does if you take out even a very small number of the worst performers.

Omaha would likely do that by knowing which business models just don't make sense, but even statistical approaches might work. It seems quite possible to identify find a chunky subset of stocks that is likely to contain a substantially-more-than-random number of big losers: you're not taking on extra risk by not merely skipping that whole subset.

It takes a very different kind of discipline to go for that kind of modest goal though. Which is why not that many folks do it.

Jim