I should have assumed that Wall Street's tendency toward reckless myopia – ingrained over the past decade – would return at the first sign of even temporary stability. The eagerness of investors to chase prevailing trends, and their unwillingness to concern themselves with predictable longer-term risks, drove a successive series of speculative advances and crashes during the past decade – the dot-com bubble, the tech bubble, the mortgage bubble, the private-equity bubble, and the commodities bubble. And here we are again.
We face two possible states of the world. One is a world in which our economic problems are largely solved, profits are on the mend, and things will soon be back to normal, except for a lot of unemployed people whose fate is, let's face it, of no concern to Wall Street. The other is a world that has enjoyed a brief intermission prior to a terrific second act in which an even larger share of credit losses will be taken, and in which the range of policy choices will be more restricted because we've already issued more government liabilities than a banana republic, and will steeply debase our currency if we do it again. It is not at all clear that the recent data have removed any uncertainty as to which world we are in.
Frankly, I've come to believe that the markets are no longer reliable or sound discounting mechanisms. The repeated cycle of bubbles and predictable crashes over the recent decade makes that clear. Rather, investors appear to respond to emerging risks no more than about three months ahead of time. Worse, far too many analysts and strategists appear to discount the future only in the most pedestrian way, by taking year-ahead earnings estimates at face value, and mindlessly applying some arbitrary and historically inconsistent multiple to them.
This is utterly different from true discounting – which does not rely on multiples, but instead carefully traces out the likely path of future revenues, profit margins, cash flows and earnings over time, and explicitly discounts expected payouts and probable terminal values back at an appropriate rate of return. That's what we actually do here. Talking in terms of multiples can make the process easier to explain, and can be a reasonable approach to the market as a whole if earnings are normalized properly, but ultimately, an investment security is a claim to a long-term stream of cash flows. It is not simply a blind multiple to the latest analyst estimate.
Fortunately, the evidence suggests that the long-term returns to a careful discounting approach tend to be strong even if investors repeatedly behave in speculative and short-sighted ways. This is because long-term returns are fully determined by the stream of cash flows actually received by investors over time, and because inappropriate valuations ultimately tend to mean-revert. In the face of speculative noise, the long-term returns from a proper discounting approach may not capture as much speculative return as might be possible, but over time, many of those speculative swings tend to wash out anyway.