Diversification doesn't work. It not only leaves your risk essentially the same in case of catastrophic market failure (see GFC/debt crisis - bonds are safe RIGHT?).
It also cuts your returns in half and brings you down along with everyone else - and gives you a false sense of security - nothing is safe - not even shorting the world (end of the world/counter party/regulatory risks).
As an investor you must think long and hard about the future/companies/investments make very few super high signal, super high impact trades/positions/companies that you know inside and out (financials/GDP winds etc).
I'd rather invest in 20 companies I know inside and out than 500 mediocre ones that I know nothing about. But that's me - most people don't know anything and they SHOULD diversify.
If you don't know anything - the best thing you could do is assume no better than random.
Local failure is obviously much more common than catastrophic failure (although the latter does have a much greater impact).
I like to think of it in terms of car crashes and catastrophes.
Companies are represented by cars - I pick good cars, with good people driving them, surrounded by a safe environment and hope they don't crash - it's how risk works.
Hurricanes destroy everyone - no matter how good the driver is - no exceptions.
The whole point of investing is to make above and beyond the S&P 500 benchmark over the long term - and that requires taking concentrated risks where one must bet on the right drivers with the right cars in awesome environments (most critical) and watching the horizon warily for distant hurricanes (debt crisis/nuclear war/WWIII/currency devaluations/governmental collapse).
I really can't do any more than that. It's a trade off obviously.
I understand local failures quite well - I rely on my ability to make decisions to mitigate that risk and the fact that I'm still young (older people should not do this).
