Bernstein on Forecasting

Bernstein wrote an editorial in the Spring 1996 Journal of Portfolio Management on forecasting that was a classic. Titled, "Fearless Forecasters, or Fearless Forecast Consumers."

After highlighting the ridiculousness of forecasting he drew the following conclusions:

(1) The question is not whether wildly wrong forecasts will happen, but what we do about the high probability that wildly wrong forecasts will happen.

(2) We should study expert forecasts as evidence of the state of expectations, but not as any kind of measure of what the future holds in store.

(3) We should concoct scenario forecasts because they force us to consider wide changes and even discontinuities.

(4) We should put more effort into managing risks and considering the consequences of being wrong.


He ends with, "pretending to believe that forecasts are going to be right may be the greatest risk of all."

There Is No Single Best Solution, But...

There is no single best investment strategy, approach, philosophy, solution. There is no unified theory of investment management.

For long term investing, there are however a few simple things you can do to increase the chance of positive outcomes.

Get on the right side of secular trends, ie. exposure to growth, increasing productivity, rising earnings/free cash flows. One important factor is the valuation level you enter at. Secondary factors are short-medium term momentum and longer term mean reversion facilitations.

Have a underlying philosophy that takes out some of the more common errors or mistakes, ie. thinking you can consistently time the market, choose the best manager, or make the right stock picks. Passive, low cost, equal weighted exposure to global growth factors.

Have the humility to realize that you don't know it all and that you won't always get it right, therefore have a disciplined downside risk management framework that either includes strict stop loss limits and/or non-correlated asymmetric downside protection.