Maybe the logical way to make this decision would be to take a look at your performance over the last decade or so and see how your picks have performed. Then if you made assumptions about
1. what percentage of your picks are losers, and what your average loss is, and
2.what percentage are winners, and what your average gain is, and
3. average amount of time it takes for market to correct to your valuation
you can get a feel for what you can do on your own vs. what the owner/manager can do.
I've also thought about this. I remember reading that Watsa said he thinks he can compound book value at 15%/yr. As long as the price/book multiple doesn't decrease, which most of you seem to think is pretty unlikely, then 10 yrs from now you can turn a dollar into four with fairfax.
I've also wondered if Watsa is being modest or intentionally setting the bar low with his 15% assumption. After all he has done quite a bit better than that in the past, albeit on a smaller base. What do the more experienced board members think about this? If you had to bet, what number would you pick for fairfax's cagr over the next decade?