Howard Marks gave a pretty awesome speech at Google. (Thanks @uptickr )
The points he makes are very interesting, and they’re all the “most important” things you should know about investing. Here’s some key parts:
You can’t tell from an outcome whether the decision taken earlier was good or not. This is from Taleb. Bad decisions get lucky all the time, good decisions fail all the tame.
You should not act as if the things that *should* happen, *will* happen. You work through probabilities instead, since many things can happen.
You can’t survive “on average”. You gotta find out the really bad days and survive on those – average makes no sense.
You don’t buy good companies at outrageous prices. You can make a good deal buying not-so-good companies at very very low prices. This is very important – because overpaying for something good is a losing proposition. Underpaying for something not good can still net you a good profit.
Consistency is more important than being on top of the charts. For 14 years a fund was always in the 27th to 47th percentile of performance in terms of comparison on a yearly basis with peers. But take the whole 14 year period, and it was in the 4th percentile. Imagine that – and it can happen because the best performers one year become the worst performers in another year because they have taken a lot of undue risks.
Passive index investing works because markets are efficient now, and people have a tough job beating the market. But it’s because they try that markets are efficient. If they stopped trying and everyone invested in Index funds, the premise of what makes markets efficient is gone. (i.e. people trying to buy stocks because they are undervalued etc. will not happen, so price discovery is gone. ) Meaning, if more people invested in index funds, markets would get LESS efficient and it will become easier to actively beat the market.
It’s a brilliant video. Even if you don’t agree with a few points, it’s a pleasure to listen to someone who’s been through at least thirty years of investment cycles.