Cliff Asness: evaluation

Transcript context

…Shotgun and can opener time, right? Yeah, exactly. Well, I use this example in a very different way. If someone says, “What if we get something five times as bad? How do you invest if we get something five times as bad as the global financial crisis?” And I say, “ammunition and canned goods.” And I don’t think there’s a better answer for that. But I do think something like coskewness — it’s a geeky idea — I think it’s very hard to establish and prove. The data is not really even there. Momentum itself, getting back to that one, it has negative skewness. The geeks call that a bad left tail. Nassim Taleb would call it a black swan event. It has standard deviations you’re not supposed to see. Big events. They have tended to be more, maybe this is luck, but they have tended to only occur in strong markets, not in weak markets. We don’t like that, but we don’t worry about that as much. To be honest, when it comes to value, Ken and Gene have never embraced this story. I don’t embrace it either, but I give it some credence. Value, buying cheap and selling expensive, has a little better of a risk story on this front, because it has suffered empirically in the Great Depression, in the global financial crisis. Probably not enough. It probably is not enough to explain it. But that is the exact kind of measure of risk that should work. Does it hurt you? This is a terrible English sentence, and I apologize in advance. “Does it hurt you when it hurts to be hurt?” is an English language version of risk. Any good quantitative measure, no matter how geeky you make it, should get back to that. If it’s a good measure of risk, it doesn’t just hurt occasionally. It hurts you when it hurts to be hurt. Are there still new and significant market inefficiencies to be found, or has that lode been mined? Are you the end of this tradition, or just the beginning?…

Stored transcript either side of the excerpt. The highlighted words are the published quote; the surrounding text is unedited source, never generated.