The Superinvestors of Graham-and-Doddsville (1984)

[1984]

I belong to a fair number of value investing forums and made money off of the strategy for a while. Here's why I think this no longer works in the traditional sense and what works now:

The rise of quants, ETFs, and instant information has largely arbitraged away value mispricings. So if it looks like a bargain, it's probably a value trap.

So where can you find value? Where the above things are not present. Quants work for big firms. Goldman Sachs can only make a dent if it does massive deals. As Buffett said, he could still do great things with $1mm AUM, but with billions, he can't make small plays anymore. So individuals can only find value where the big players (GC, ETFs) don't play. This means micro caps, foreign stocks (Japan comes to mind as a hot spot for value). The problem is two fold:

1. The above-mentioned pool is very small.

2. It's riskier.

So you have to do a ton of research to avoid the value trap mistake, often with way less information since these stocks aren't subject to the same 10K/10Q auditing that American stocks are.

Now you've researched something so much you're biased to believe it working since you've sunk so much time into it. And because you've sunk so much time you don't have the time to research the rest of the investment pool, so combined you see these value investors who are very concentrated in some highly-convicted bets. And thus, the ones that win, win big and can claim there is always value to be found even in a market dominated by momentum investing. The rest lose to value traps, and lose big.

So what's the takeaway? It still works. The low numbers in terms of P/E and P/B that are in books like The Intelligent Investor don't work. You have to relax those constraints quite a bit. And you can't be looking in the S&P 500.

I used to think I could do this as a hobby. And I did. But I think I got lucky based on the amount of time and research I did. To be demonstratively good at value investing time and time again requires robotic levels of dispassionate patience, and research that demands a full time job.

I'd still argue than metrics like EV/FCF give some indication of valuation / risk.

Balance sheets become less meaningful going forward. Google's greatness is not to be found in its physical assets.

Buffett's thinking is nearly always a pleasure to read for its clarity. If you'd read that in 1984 and put your money into Berkshire on that basis you'd be up about 250x.

Yes, I agree! Metrics are extremely important in screening out for value stocks. What I'm saying simply is that there is no singular magical metric that can wipe out hours and hours of research. To simply list all stocks by EV/FCF ordered by cheapness will not automatically create a winning portfolio.