Harvard and Yale concentrated with venture capitalists that got the best calls and brainpower. Very few firms made most of the money, and they made it in just a few periods. Everyone else returned between mediocre and lousy. When returns happened, envy rippled through institutional money management. The amount invested in venture capital went up 10 times post-1999. That later money was lost very quickly. It will happen again. I don't know anyone who successfully resists this stuff. It becomes a new orthodoxy.

Accounting is a big subject and there are huge forces in play. The entire momentum of existing thinking and existing custom is in a direction that allows terrible follies to happen, and the terrible follies have terrible consequences.

Berkshire's whole record has been achieved without paying one ounce of attention to the efficient market theory in its hard form. And not one ounce of attention to the descendants of that idea, which came out of academic economics and went into corporate finance and morphed into such obscenities as the capital asset pricing model, which we also paid no attention to. I think you'd have to believe in the tooth fairy to believe that you could easily outperform the market by seven-percentage points per annum just by investing in high volatility stocks.

Where you have complexity, by nature you can have fraud and mistakes. You'll have more of that than in a company that shovels sand from a river and sells it. This will always be true of financial companies, including ones run by governments. If you want accurate numbers from financial companies, you're in the wrong world.

So, economics should emulate physics' basic ethos, but its search for precision in physics-like formulas is almost always wrong in economics.

The stupid and dishonest accountants allowed the genie of totally inappropriate accounting to descend on derivatives books. And once this has happened - people get status, etc. - it's impossible to get it back into the bottle.

After nearly making a terrible mistake not buying See's, we've made this mistake many times. We are apparently slow learners. These opportunity costs don't show up on financial statements, but have cost us many billions.