Love this:
"Periodically, financial markets will become divorced from reality – you can count on that. More Jimmy
Lings will appear. They will look and sound authoritative. The press will hang on their every word. Bankers will
fight for their business. What they are saying will recently have “worked.” Their early followers will be feeling very
clever. Our suggestion: Whatever their line, never forget that 2+2 will always equal 4. And when someone tells you
how old-fashioned that math is --- zip up your wallet, take a vacation and come back in a few years to buy stocks at
cheap prices."
That's the part I would have highlighted too, but because I think it's the worst part of the whole letter! Or at least the last line is. I think he temporarily forgot that almost no one reading his letter is named Warren Buffett, and accidentally deviated from his normal advice.* While market-timing might work great for him, it doesn't for anyone else!
* Luckily he explicitly contradicts himself elsewhere in the letter:
"Investors, of course, can, by their own behavior, make stock ownership highly risky. And many do. Active trading, attempts to “time” market movements, inadequate diversification, the payment of high and unnecessary fees to managers and advisors, and the use of borrowed money can destroy the decent returns that a life-long owner of equities would otherwise enjoy. Indeed, borrowed money has no place in the investor’s tool kit: Anything can happen anytime in markets. And no advisor, economist, or TV commentator – and definitely not Charlie nor I – can tell you when chaos will occur. Market forecasters will fill your ear but will never fill your wallet."
I don't think it's a contradiction, he's talking about different kinds of phenomena. In the first, he is essentially saying that someone, somewhere will always invent a new get-rich-quick scheme, and that people will tend to ignore obvious risks and latch onto it. In the 80s, this was LBOs where you put down $5 of equity to buy a $100 business, and people were happy to lend you 95% of the purchase price. In the 90s, it was internet companies with no revenue and crazy business models. In the 2000s, it was literally the idea that you could have 2+2=5, that you could package mortgages and other debt in ways that virtually eliminated the underlying risks.
In the 80s, in the 90s, and in the 2000s, you did well by buying a diversified set of stocks like the S&P, because regardless of the above phenomena, the economy and businesses in general will do better over time. Even if you observed the above phenomena, generally it would have been a bad decision to time the market; for example, you could have observed the craziness of the Dot-Coms in 97, and you would have done worse than someone who just held the S&P the whole time, even considering the crash in 2000-2002. Unless you timed the real estate bubble perfectly by selling in 07 and buying in 09, you probably did poorly or at least no better than the market.
The point being, micro-level events that are crazy and irrational are bound to happen in financial markets, and you should zip up your wallet when confronted with opportunities that don't make sense. But that doesn't mean you can effectively time the market by buying and selling an index based on the activity you see.