I’ve posted Entry #278 to my weekly Valuation-Informed Indexing column at the Value Walk site. It’s called Our Investing Biases Are Particularly Dangerous Because They Are Time-Based Rather Than Phenomenon-Based.
Juicy Excerpt: The most important investing biases are time-based rather than phenomenon-based. That means that for long periods of time certain ideas are forgotten by almost the entire population. To tap into the other side of the story the investor would have to study historical data from a time-period many years removed from the current time-period. Who does that?
Shiller showed that valuations affect long-term returns. What he really was doing when he did that was showing that the stock market is not efficient, that mis-pricing on either the high or low side is a significant reality rather than the illusion that Buy-and-Holders believe it to be. Even during the most out-of-control bull market, there are a small number of people questioning whether the insane prices achieved are real and lasting. But the percentage of the population holding that view can be very small indeed. The percentage of the population that is conservative rather than liberal doesn’t vary dramatically from time to time. The percentage of the population that believes that stocks are the perfect investment choice is dramatically higher when prices are high than it is when prices are low.
For a good number of years following the great crash of 1929, investors didn’t expect to see any capital appreciation at all on their stocks. The conventional wisdom of the time was that stocks were worth buying only for their dividends; those that didn’t pay high dividends were not worth owning. In the late 1990s, dividends fell to tiny levels. The very thing that made stocks dangerous (their high price) changed the conventional wisdom on stock ownership to reflect a bias that stocks are always worth owning.
Stocks for the Long Run was a popular book in the 1990s. It would not have sold many copies in the 1930s. The book reports on data, facts, objective stuff. The message of the data should not change from times like the 1930s to times like the 1990s. But the ways in which we arrange the data and interpret the data changes when we go from bull markets to bear markets. People will be looking at the same data that was employed in Stocks for the Long Run to sell stocks to make the case against stocks when we are on the other side of the next stock crash.